Document 6mJ3bmBxoxw9Y0dVQ4xpBkzg

LYMPHADENOPATHY INDUCED BY ANTICONVULSANT DRUGS AND MIMICKING CLINICALLY AND PATHOLOGICALLY MALIGNANT LYMPHOMAS SIDNEYL. SALTZSTEIN, M.D.,+ AND LAURENV. ACRERMAN, M.D. SINCE THE introduction of the various hy- tions, the serious side effects, and the availdantoin derivatives and analogues into ability of other means of treating Sydenham's the therapy of convulsive di~orders,~scQattered chorea, Nirvanol has never been an approved reports have appeared in the literature of a drug.16 peculiar lymphadenopathy directly related to I n 1940 &ope and Burrows16 reported on this therapy. These reactions closely mimic a patient (Table , case 1) who developed a various malignant lymphomas, both clinically severe macular eruption 1 week after the start and pathologically, and a diagnosis of lym- of diphenylhydantoin (Epanutin) therapy. phoma might be made by the unwary physi- This was accompanied by pyrexia and palpa- cian. It is for this reason that this report is ble nodes in the posterior triangle of the neck. being published. Most of the anticonvulsant Evidently the symptoms disappeared with ces- drugs have been implicated in this reaction, sation of therapy, as the authors stated that with methoin (Mesantoin) being most fre- there was a recurrence when an attempt was quently implicated. The lymphadenopathy made to resume treatment. The report of may occur in conjunction with the more Harrison et al.,s4 in 1946, of the first fatality widely known reactions to these drugs, such as following Mesantoin therapy stated that the leukopenia, fever, and skin rashes, or it may patient had slight cervical lymphadenopathy. appear in the absence of these other reactions. This patient (Table 1, case 2) was receiving Only a few of the previous case reports in- diphenylhydantoin (Dilantin), trimethadione cluded any histological description of the (Tridione), and Mesantoin, and she died of lymph nodes. In the past 5 years, 7 individuals aplastic anemia. At autopsy, the lymph nodes whose clinical courses resembled those previ- had small germinal centers and showed extra- ously described have been admitted to either medullary hematopoiesis. Barnes Hospital or St. Louis Children's Hos- Gaustad, in 1947,Zr described cervical nodes pital, St. Louis, Mo., and biopsies of appro- that were slightly enlarged and tender in a priate tissues have been performed. These pa- 27-year-old man (Table 1, case 3) who had tients form the basis of this report. been receiving Hydantal(3-methyl-3,Fi-phenyl- ethylhydantoin) for 25 days. The adenopathy REVIEWOF THE LITERATURE was associated with a morbilliform-to-macular rash, fever, leukopenia, and malaise. At about In the late 1920's and early 1930's, when the same time, a preliminary report of 2 pa- Nirvanol(5-ethyl-5-phenylhydantoinw)as used tients with cervical adenopathy in the absence in the treatment of Sydenham's chorea, lym- of blood dyscrasias appeared.41 These cases phadenopathy was a frequent accompaniment were evidently later incorporated into the re- of the intentionally induced drug reaction port of Bercel et al.7 Since then several other that was the goal of the therapy.88*50*52.5638*, cases have been reported, and they are sum- 5 9 1 6 6~3 However, no reports of the histology of marized in Table 1. the enlarged nodes are available, and no seri- I n 1948 3 reports appeared with gross ous consequences of the adenopathy itself are and/or microscopic descriptions of the in- reported. Because of the severity of the reac- volved nodes. Van Wyk and Hoffman64 de- scribed the case of a 71-year-oldman (Table 1, From the Division of Surgical Pathology, Washington University School of Medicine and the Barnard Free Skin and Cancer Hospital, Barna Hoapital, St. Louis, Mo. Present address: Armed Forces Institute of Pathol- k rogy, Washin ton, D.C. Received publication M a y 5, 1958. case 10) who had been receiving Dilantin for about 5 months. He entered the hospital with stomatitis and exfoliative dermatitis. Enlarged left cervical and inguinal nodes were noted on (Text continued on page 168.) TABL1E CLINICAL AND PATHOLOGICAL SUMMARY O F REPORTED CASES OF ANTICONWLSANT-INDUCED LYMPHADENOPATHY Case no. 1 2 Ref. no. Yr. of ref. 16 1940 34 1946 3 27 1947 41 1947 Age Drug Race Durat. Sex admin. Nodes involved . . . Epanutin Cerv. 1 wk. .1.6. Dilantin, Tridione, Cerv. (slt.) F Mesantoin Other symptoms Fever, rash Aplast. anemia, 1% eosin. 27 ~ y ; l & t a ~ Cerv. M' 3 wk. (tend.) Fever, rash, leukopenia Pathological findings Follow-up ... Recur. when attempt was made to resume treat. Autop.; germi. Died cent. small, extra- med. hema- topoiesi.s.. Cleared with cessat. treat. ... Mes.a.n.toIin Cey. (1"sue 17 1949 goose . egg") 2 with leukopenia ... 2 with leukopenia recov. when drug discont.; other 3 "reces. & redup. respect. on withdrawal &readmin. ther." 1948 .9 10 1948 11 ... Tridi.o.ne Cery., Ingum., Rash, Autop.; mesen- Died agranulocyt. teric nodes & F axil. Peyer's patches hyperplast. (no microscop.) 10 64 71 Dilantin Cery., Anemia, Biop.; destruct. Died; autop.; periarter. nod- 1948 N 5 mo. Ingum. exfol. architect., endo- osa involv. liver, spleen, M dermat. thel. prolif., infil- kidneys, skin, bone marrow trat. eosin. ...11 29 1948 w61 Dilantin, Tridione Few slt. enlarged Granulocytopenia Autop.; nodes slt. Died enlarged grossly, M Clin. unremark. 12 35 1948 .1.7. Mesantoin 1st mo. Gen. Fever, microsc.o.p.. pharyngitis Cleared with cessat. ther.. did not recur with resumpt. F ther. w13 28 11 Tridione, Cerv. 1948 diphenyl- (slt.) Fever, Autop.; (no Died pancytopenia mention nodes) 21 .F. . ene 2 yr. Mesantoin Gen. Rash, feyer, ... Cleared in 1-2 wk. after 1948 4-8 wk. (most 15% eosin. cessat. ther. 22 prom. 16 17 1949 48 ... cerv.) Mesantoin Not Fever Biop.; "indis- Recovered on withdrawal 1949 2 wk. specif. tinguish. from drug; fever, rash & adenop. Hodgkin's dis." on taking drug again; re- ...18 42 1949 .6.9 Tridione 6-7 wk. Cerv., submax., Rash, icterus, 26% covered again on withdrawal Cleared in 5 wk. on Cessat. drug M' axil. eosin., hepa- tosplenomeg. 19 4 29 Mesantoin Gen. Chills, fever, ... Cleared in 2 wk. on cessat. 1949 6 d a y s (espec. rash, leuko- drug 'F' cerv.) penia, hepa- 20 8 30 Mesantoin Not tosplenomeg. Rash, fever, ... Dose reduced briefly & then 1950 W 1 wk. specif. pharyngi tis reinstated with exacerbat. M toxic manifest.; after 1 mo., grad. inbeas. doses without Thyphenyl- Cerv. Fever, ... toxicity Recovered with cessat. drug toin (in pharyngitis, add. to pancytopenia Mesantoin) . . .21 47 ... 3 mo. Mesantoin Cerv. Rash, 1950 pharyngitis, ... "Therapeutic test reproduced path. picture" tonsillitis 22 14 29 Dilantin Cerv. Rash, fever, Biop.; "chronic Cleared with cessat. drug; 1950 N 17 days axil., icterus, 6%- inflam." trial dose 5 wk. later caused M inguin. 20% eosln., rash & eosin.; cleared on (tend.) hepatomeg. cessat. ther. TABLE1 (Continued) CLINICAL AND PATHOLOGICAL SUMMARY OF REPORTED CASES OF ANTICONVULSANT-INDUCED LYMPHADENOPATHY Ref. no. Age Drug Case Yr. of Race Durat. no. ref. Sex admin. Nodes Other involved svmotoms Pathological findines FO~~OW-UD 23 56 13 Mesantoin Gen. Splenomeg., . . . Died 1950 . 3mo. rash, M' 26 1950* ... ... Mes.a.n.toin Not specif. pancytopenia "SeV.,, react. , ... Cleared with cessat. ther. 26 fever, pharyngitis, -27 37 1951 ... ... Methoin 4 mo. Cerv. pancytopenia Rash. malaise, ... Returned to norm. after drug stopped M lymphocyt., fati ue 28 15 21 Mesantoin Cerv. Ra&, 6% Biop.; destruct. Resolved with smaller doses; 1951 . 3wk. (sev.) eosin., architect., retic. recurred in 17 days, disap- M. pharyngitis cell hyperplas., pear. rapid. when pt. shifted mitoses, necrosis, to Luminal phagocyt., eosin. 29 15 1951 .4.7. Mesantoin Gen. 3 tab. M only Rash, fever, Biop.; similar to Resolved with cessat. ther.; bronchitis case 28 with less rea-m-ear. when drug restart. I necrosis, retic. cell hyperplas., mitoses, 30 36 -19.51 29 W Mes.a.n.toin Inguin. Skin pigmenta., phagocyt..., eosin. F 45 1952 ...t Mesantoin 4mo. Not specif. 3% eosin. Fever, rash over all ... ... 44 or less nodes, hy- pertrichosis, skin pigmen- tat., blood dyscrasias in some nodes 45 51 36 Mesantoin Cerv., Fever, ... Regressed slowly on cessat. 1952 . 20days submax., myalgia drug F' mediast. 46 51 56 Mesantoin Cerv., Fever, Biop.; architect. Disappear. in 1 mo. when 1952 . 5wk. then gen. urticaria, altered, retic. drug stopped M 11% eosin. h perplas., many p L m ? cells, necrosis 47 51 1952 .31 Mesantoin Cery., 1 mo. inguin. Rash, slt. Biop.; alterat. Sympt. re essed on hepatomeg. struct., retic. cell stopping cKug F ' hyperplas., eosin., degen. 48 51 46 Mesantoin Cerv., Prev. hist. Biop.; norm. Persist. fever, renal fail., 1952 . 3mo. M +I.,, inguin. skin erupt., follic. pattern diarrhea, death; drug evigrippelike with pleomorp. dent. contin. to death symp., bron- react., no necro- chitis, fever, sis, h perplas. hepatomeg. lyrnpioid & retic. elements & endothel. . . .49 62 11 Dilantin Cerv. Fever, rash, Recovered rapidly on cessat. 1953 W 2 w k . (slt.) 6% eosin. ther. M 50 46 1954 . .5. Dilantin & phenyl- All palp. nodes, Fever, F h , Biop.; retic. cell Recovered with cessat. ther., 2y0 eosin. hy plas. replac. nodes recur. with resumpt. M acetylurea rnediast. folEes, sclerosis & disappear. again with ces- . . .19 ... 1 mo. (Xray) Mesantoin Not at hilum Rash, fever, I.. 1954 specif. blood alterat., sat. ther. ... .he to- sp E o m e g *The report by h i ! and G i b W included 18 patients who received hydantoin therapy and in whom node involvement was "usual. Other symptoms included rash, fever and pharyngitis; these regressed in 4 to 7 days whteSn ithoef rtahpeysew1a4s stopped. patients were children, and 8 were adults. TABLE1 (Conclded) CLINICAL AND PATHOLOGICAL SUMMARY OF REPORTED CASES OF ANTICONWLSANT-INDUCED LYMPHADENOPATHY Ref. no. Age Drug . . . ... 14 pt., 72 inguin. & submax. in 2 pt. 73 Fever, rash, icterus, hepatomeg., 35% eosin. 74 43 1957 .25 Mesantoin Cerv., 18 mo. axil. Clin. diag. dissem. lup. ... F erythem. 75 43 18 Mesantoin Cerv., Fever, rash, 1957 - . 1 mo. axil., 7y0eosin., F' pos. lup. erythem. test Case 1 11) Pres. W 6wk. "Acute Append.; tum. Recovered rapidly: well 5 yr. adenop., abdomen" retx. cells with postop. ser. M mesenteric num. mitotic nodes en- figs., eosin., foci larged at necrosis & 0 phagocyt. 77 Case 2 Pres. 6 Mesantoin &y, W lwk. inguin., Rash, conjunct., Biop.; retic. cell Recovery with cessat. & lymphoid Mesantoin ther.; seiz. being ser. M axil. 3%-9% eosin. hyperplas., control. by Gemonil architect. pre- served, few eosin. & sma!l foci necrosis Plasmocyt. Biop.; lst, some Nodes disappear.with cessat. lo%,abnorm. dace. norm. Mesantoin; reappear. & dis- inguin. gen. ser. prot. architec., hyper- appear. with resumpt. & (no diag. plas. retic. ele- cessat. Mesantoin; still has mult. myel.) ments, few plasmocyt., & abnorm. ser. plasma cells & prot. eosin.; 2d, re- place. most norm. architect., retic. cell hyperplas., mitow?, fewareas necrosis, eosin. Biop.; lst, much obliterat. normal architect., retic. cell hyperplas., mitoses, nuclear debris, eosin.: 2d, sim. to 1st with more w i n . Biop. ;hyperplas. lymphoid ele- ments with slt. incr. in retic. cells Case 6 7 Peganone Cerv., PreS. W 5wk. axil., Biop.; compl. Nodes returned to norm. 'SpGnomeg., obliterat. norm. within days cessat. ther. ser.$ M 4%-15% architect.; infiltrat. lymphocytes, retic. & plasma cells, eosin.; necrosis & Case 7 Pres. ser. 45 Dilantin N 2wk. F Gen. Rash,fever, m e i s e , 5% eosin., icterus, &em. evid. liverdamage - $This case has been reported p r e v i o u ~ l y . ~ ~ Recovered rapidly with destruct. nodal cessat. ther. architect. with infiltrat. pleomorp. retic. cells, playna cells, emin. 168 CANCEJRanuary-February 1959 VOl. 12 admission. An inguinal node was biopsied, Several other articles have appeared in the and the histological examination revealed de- European literature. Chiari15 reported 2 cases struction of the normal architecture, endothe- in 1951. The first (case 28) was of a 21-year- lial prominence with capillary proliferation, old man who had been receiving Mesantoin and eosinophilic infiltration. Pigmented mac- for 3 weeks when he developed a sore throat, rophages were also found in the node, and the a morbilliform rash, and cervical lymphade- interpretation at the time was a reactive lym- nopathy. He had 6% peripheral eosinophilia. phadenitis secondary to the exfoliative derma- A lymph node was biopsied and showed de- titis. At autopsy, 4 days later, the lesions of struction of most of the normal architecture. periarteritis nodosa were found in the liver, Reticulum cell hyperplasia with occasional spleen, kidneys, skin, and bone marrow. mitotic figures was noted. Areas of necrosis Gentry and Hill20 reported the case (Table and phagocytosis of cellular and nuclear de- 1, case 11) of a 6S-year-old boy who died of bris were prominent, and eosinophils were granulocytopenia, which they attributed to numerous. The patient's therapy was changed Tridione therapy. He had a few slightly en- to small doses of Mesantoin with resolution larged nodes on admission, and at autopsy the of the lymphadenopathy. Seventeen days later, nodes were also described as being slightly en- however, the nodes reappeared. They subse- larged. Microscopically the nodes were "unre- quently disappeared rapidly when the medica- markable." Braithwaitelo described a similar tion was again changed to Luminal. Chiari's case (Table 1, case 9), this of an ll-year-old second case (case 29) was that of a 47-year- boy who died of agranulocytosis while being old man who had received only 3 tablets of treated with Tridione. On admission, the cer- Mesantoin. He developed a rash, bronchitis, vical, axillary, and inguinal nodes were en- and lymphadenopathy. His medication was larged. At autopsy, the mesenteric nodes were changed to Luminal with resolution of symp- also found to be enlarged, and Peyer's patches toms. Mesantoin was tried again, and 2 days were described as "hyperplastic." There was later the node swelling reappeared. On the no reported microscopic examination of these following day the node was biopsied. It nodes. A lymph node biopsy, which was mic- showed reticulum cell hyperplasia with many roscopically "indistinguishable from those ob- mitoses. Cellular degeneration, phagocytosis, tained in Hodgkin's disease" was reported by and eosinophilic infiltration were noted. Some Meller and Resch48 in 1949. The patient lymphoid follicular hyperplasia and sparse (case 17) had been on Mesantoin for 2 weeks plasma cells were also present. and also had fever. T h e symptoms abated Olmer et a1.,61 in 1952, described lymphade- when the drug was stopped. T h e patient was nopathy in 4 people (cases 45 to 48) receiving given Mesantoin again and developed fever, Mesantoin. I n 3 of these, nodes were biopsied rash, and adenopathy; again there was re- and showed similar changes. Reticulum cell covery when the therapy was stopped. hyperplasia and plasmocytosis were present in In 1950 Chaiken et al.14 reported the case all 3, and necrosis and infiltration with eosin- (case 22) of a 29-year-oldman who, after being ophils were present in 2. Color photomicro- on Dilantin therapy for 17 days, developed graphs of the findings were included in the malaise, myalgia, a pruritic rash, icterus, and article. Cytological changes similar to all of fever. On admission, there were large, tender those previously mentioned were reported by nodes in the cervical, axillary, and inguinal BodartQ in 1953. I n 1954, Martin et aL46 regions. The liver was also enlarged. T h e described a 5-year-oldboy (case 50) who devel- white blood cell count was 8,000, with eosino- oped a fever and rash 10 days after the initia- phils ranging from 6% to 20%. Bone marrow tion of therapy with Dilantin and phenylace- examination revealed an eosinophilia ranging tylurea. The drug therapy was stopped, and from 28% to 57%. A cervical node was b i o p there was a remission of symptoms. T h e ther- sied and disclosed "chronic inflammation." apy was restarted, however, 10 days later be- No arterial lesions were described in a simul- cause of recurrence of the seizures. Four days taneous muscle biopsy. T h e patient's signs later the boy again developed fever and rash, and symptoms cleared with cessation of Dilan- this time accompanied by enlargement of all tin therapy. Five weeks later, a trial dose of the palpable lymph nodes as well as enlarge- the drug was instituted, and the malaise, rash, ment of the mediastinal lymph nodes (as dem- and eosinophilia reappeared. These again onstrated on roentgenograms). There were subsided with cessation of therapy. only 2% eosinophils in the peripheral blood. No. 1 -DRUGINDUCLEYDMPHOMA-LAIDKEENOPATHY Saltzstein 6.Ackerman 169 The sternal marrow was normal. An inguinal lymph node was biopsied and showed reticulum cell hyperplasia replacing the follicles and sclerosis at the hilum. There was no mention of eosinophils or necrosis. The child recovered with cessation of the drug therapy. The most recent report is that of Lindqvist,M in 1957, who described 2 cases of disseminated lupus erythematosus after Mesantoin therapy. His second case (Table 1, case 75), was that of an 18-year-old woman who developed fever and cervical lymphadenopathy 1 month after Mesantoin was added to the Epanutin she had been taking. Without any change in therapy, her fever would periodically decrease and increase. After about 6 or 8 weeks, she developed a persistent fever, generalized nontender lymphadenopathy, and a "butterfiy" rash that eventually became generalized. Lupus erythematus cells were found in the blood. An axillary lymph node biopsy was performed and showed nonspecific lymphadenitis with an increase of eosinophilic ...cells. Lindqvist felt that "the histologic pic- ture tallied rather well with that described by [Chiarila and Olmer et al.611, although the changes were slightly less pronounced." The patient's symptoms, including the lymphadenopathy, responded to adrenocorticotropic hormone therapy. Several patients included in Table 1 showed only minimal lymphadenopathy. T h e patient of Harrison et al.34 (case 2) showed only slight cervical lymphadenopathy, but there was a reduction in the size of the lymph node germinal centers at autopsy, along with extramedullary hematopoiesis. T h e nodes in the patient reported by Gaustad27 (case 3) were also only slightly enlarged, but they were tender. T h e child described by Gentry and Hill29 (case 11) also had only slightly enlarged nodes, both clinically and at autopsy. In the second case described by Hunter and Jenkins86 (case 30) there was inguinal lymphadenopathy along with 3% eosinophilia. This is not a significant degree of eosinophilia, and inguinal adenopathy is at best difficult to evaluate. However, no such adenopathy was described in relation to their other 4 patients. Steinberg's62 patient (case 49) had only slightly enlarged cervical nodes, but he did have 6% eosinophils in the peripheral blood, along with rash and fever, and he recovered rapidly on cessation of ther- apy. Both the patient of Gayle et a1.28 (case 13) and Lindqvist's4Sfirst patient (case 74) showed latent periods much out of proportion to the rest of the patients in this series, and both had only slight cervical lymph node enlargement. The inclusion of these 2 patients in this series might not be justified. DeJong17 described a case, not included in this series, of jaundice resulting from phenylacetylurea (Phenurone) therapy. A previous diagnosis of infectious mononucleosis had been made, suggesting that there might have been lymphadenopathy. Both the report of Carnicelli and Tedeschi18 and that of Forster et al.24 mentioned changes in lymph nodes as results of fatal bone marrow depression, but they did not describe either clinical lymph node swelling or the changes in nodes that others have described in relation to anticonvulsant therapy. Several review articles also mentioned lymphadenopathy related to anticonvulsant therapy, but they did not cite specific examples.6, 6.26. 60 There is only one report of experimental production of lymphadenopathy by anticonvulsants. Kaslaris8Q administered Mesantoin to 4 kittens, 1 of which developed slight lymph node enlargement that on section showed hyperplasia of reticulum cells. Necrosis and mitoses were not present. T h e nodes in the other 3 kittens were histologically normal. Interestingly, the gingival hyperplasia associated most often with Dilantin but also with the other drugs has been produced in ferrets.1 ' CASEREPORTS Case 1. I n January, 1948, the case of a white boy, 6.5 years old, was diagnosed as that of a behavior problem. He started to have petit ma1 seizures in June, 1950, and was tried on a variety of drugs, including Tridione, paramethadione (Paradione), phenobarbital, and D-amphetamine (Dexedrine), with only slight improvement. In October, 1952,the Paradione was discontinued. Phensuximide (Milontin) was first given on Oct. 21, 1952, and increased gradually to a maximum dose of 2.7 g m . per day. For 5 days prior to admission, which was on Dec. 12, 1952, he experienced midabdominal pain without nausea or vomiting. On the day of admission, he developed vomiting and diarrhea. On physical examination at the time of admission to St. Louis Children's Hospital (on the service of J. Hollowach), there was tenderness and spasm in both the right upper and right lower quadrants. There was a suggestion of rebound tenderness. N o cervical nodes were palpable. Laboratory data showed a hemoglobin level of 13.2 gm. and a white 170 CANCERJanuary-February 1959 VOl. 12 FIG.1. Case 1. (WUIll. 58-1100A).Section of the appendiceal mass showing pleomorphic reticulum cells, eosino- phils, neutrophils, and nuclear debris. (H. & E. X750.) blood cell count of 9,500. Segmented leuko- cytes represented 91% of the differential count, and eosinophils ranged from 1% to 2%. A preoperative diagnosis of appendicitis was made. He was operated upon (by C. B. Mueller), and several enlarged mesenteric lymph nodes were found. The appendix was intussuscepted into the cecum, and at the base of the appendix a mass of soft, red tissue was noted. The appendix and the mass were removed. Postoperatively the patient recovered rapidly. Grossly, the appendix was congested, but it was not necrotic or perforated. At its base, it was surrounded by a mass of what ap- peared to be lymphoid tissue, 1 cm. in diameter and 3 ml. in thickness. Microscopically. the architecture of the proximal portion of the appendix was obliterated, and the lumen was occluded by a cellular tumor composed of reticulum cells with numerous mitotic figures (Fig. 1). Eosinophils were prominent in patches and lymphocytes were present, but Reed-Stemberg cells were not seen. Foci of necrosis with phagocytosis were common. T h e distal portion of the appendix was acutely inflamed. The pathological impression at the time, and on repeated review, was of malignant lymphoma. A subsequent bone marrow study was diagnostically nonspecific. The Milontin therapy was discontinued when the patient was discharged, and he was maintained in phenobarbital. I n October, 1953, because of grand ma1 seizures, Dilantin was added to the therapeutic regimen. A barium enema and upper gastrointestinal roentgenographic series on Oct. 6, 1954, showed no abnormalities. Acetazolamide (Diamox), in a dose of 250 gm.4 times a day, was added to the phenobarbital and Dilantin therapy in July of 1956. I n March, 1957, an attack interpreted as viral hepatitis, associated with light stools, dark urine (bile and urobilinogen present), jaundice, and minimal splenomegaly, resulted in cessation of the Diamox therapy. He recovered rapidly. A barium enema at that time was entiCely normal. When last seen, in November, 1957, 5 years after the operation, he was in good general health. A few small inguinal and cervical nodes were palpable, but the liver was not felt, and only one observer thought he felt the spleen at the left costal margin on unusually deep inspiration. T h e white blood cell count was 3,700 with a normal differential. At present, he is being maintained on only phenobarbital and Dilantin therapy. Case 2. A white boy, at 16 months of age, No. 1 DRUGINDUCELDYMPHOMA-LAIDKEENOPATHY Saltatein 6.Ackerman 171 had fallen from a second-story window and suffered a cerebral concussion. He had one clonic convulsion shortly after the fall. He a E4parently recovered completely, but 6 mont later he began having generalized convulsions and petit ma1 seizures. Dilantin and phenobarbital were, started at that time (April, 1949). The child had a history of allergic rhinitis. In January, 1954, he started to have in- termittent periods of fever, which continued until his hospital admission. On March 29, 1954, 0.1 gm.per day of Mesantoin was added to his anticonvulsant regimen. One week later, he developed lymphadenopathy, a morbilliform rash, and conjunctivitis. The adenopathy continued until the time of hospital admis- sion. Nine days before admission, his fever recurred, and 7 days before admission, the lymphadenopathy became more marked. He was I.admitted to St. Louis Children's Hospital (on the service of Hollowach) on May 17, 1954. T h e only significant physical findin s on aadmission were bilateral cervical lymp ade- nopathy, consisting of several nontender, nonmatted, freely movable nodes, and smaller inguinal and axillary nodes. T h e admission laboratory data showed a white blood cell c a n t of 8,200. The differential count included eosinophils ranging from 3% to 9%, a neutro- phi1 count ranging from 31% to 71%, and a lymphocyte count ranging from 39% to 58%. T h e hemoglobin level-was 11.7 gm. A bone marrow study on May 18 was diagnostically nonspecific. On May 20, a cervical lymph node was resected. Four days later, a tonsillectomy and adenoidectomy were performed. Unfortunately, the tissue from the latter operation was not examined histologically. The lymph node measured 8 ~ 4 x 5ml. and had a pinkyellow color `on section. Microscopically, there was slight hyperplasia of both lymphoid and reticulum cells, but the normal architecture with germinal centers was preserved. A few eosinophils, some small foci of necrosis, and a rare mitotic figure were present (Fig. 2). No Reed-Sternberg cells were seen. T h e Mesantoin therapy was discontinued while the boy was still in the hospital. His seizures have since been controlled by methbarbital (Gemonil). phenobarbital, and Dilantin. He has remained in good general health. When last seen, in February, 1958, no nodes were palpable. Case 3. A 53-year-old white woman was ad- mitted to Barnes Hospital (on the service of R. Steinkamp) on July 18, 1954, with a chief complaint of painful lumps on her body. She FIG.2. Case 2. (WUIll.58-1101A).Cervical lymph node showing one of the-kw ar& of pleomorphic reticulum cell hyperplasia with nuclear debris. (H.& E. x750.) 172 CANCEJRanuary-February 1959 VOl. 12 FIG. 3. Case 3. (WU Ill. 58-1102A). Inguinal lymph node showing eosinophils, pleomorphic reticulum cells, lymphocytes. plasma cells, and nuclear debris. (H. & E. ~750.) had been receiving phenobarbital for 5 years as therapy for mild hypertension. In Novem- ber, 1953, she had had her first convulsive seizure, and she had had 2 more, in February and March, 1954, respectively. Nicotinic acid therapy was started in March. In early April, an electroencephalogram was interpreted as showing a convulsive disorder, and Dilantin therapy was started. Later in April, approximately 11 weeks before her admission to the hospital, she noted a painful, hard lump under the right mandible. This was not altered by penicillin therapy, and subsequently a similar nodule appeared under the left mandible. On June 1,one of these nodes was removed and showed hyperplasia of the reticulum elements. There was some effacement of the normal architecture, but germinal centers were still seen. Lymphocytes were noted growing out into the capsule. A few plasma cells and eosinophils were seen. I n the ensuing 6 weeks, the patient developed tender masses in the right axilla and left inguinal region. She also had other tender areas over her body. There was no fever. On July 12, a bone marrow aspiration was unremarkable except for plasmocytosis of 10%. On admission, smooth, slightly tender, en- larged lymph nodes were noted bilaterally in the submandibular regions, in the left cervical and inguinal groups, and in the right axilla. The hemoglobin level was 13.7 gm.,and the white blood cell count was 5,900 with only 1% eosinophils. A heterophil antibody titer was negative, and serum protein electrophoresis was interpreted as "abnormal beta-I globu- lin.'' T h e clinical impression at the time of admission was multiple myeloma (or other lymphoma), but the possibility of a drug re- action was entertained. On July 21, an inguinal node was removed. It showed focal hemorrhage grossly. Microscopically, most of the normal nodal architecture was replaced by a mixed reaction of reticulum cells (often with mitotic figures), plasma cells, eosinophils, and fibroblasts (Fig. 3). Some areas of necrosis were seen, and the capsule did contain lymphocytes. N o Reed-Sternberg cells were found. A drug reaction was among the pathological diagnoses considered at the time. Dilantin was discontinued on July 23, and the patient was maintained on phenobarbital. The lymphadenopathy disappeared. Subse- quent to this a trial dose of Dilantin was given, with the reappearance of the nodes. They subsequently disappeared again when the drug was stopped. At present, the patient is alive and well, but she still has the plasmo- No. 1 DRUG-INDUCLEYDMPHOMA-LAIDKEENOPATHY Sulkstein 6.Ackerrnan 173 cytosis and abnormal serum proteins. There is no other evidence of multiple myeloma. Case 4. A 26-year-old white male sheet metal worker was admitted to Barnes Hospital (on the ward service) on Sept. 14, 1956, with a chief complaint of "swellings" on his body for 4 months. When he was 2 years old, he had fallen down a flight of stairs. Shortly there- after he had developed "blanking-out" spells, and eventually a motor component appeared in the seizures. For 6 to 7 years, he had been taking Dilantin. Early in April, 1956, because of the increased frequency of his seizures, Mesantoin (1 tablet twice a day) was given in addition to the Dilantin. Later in April, he had an episode of fever and weakness associated with a macular rash. This was interpreted as tonsillitis or scarlet fever, and he fwas 'ven some pills that possibly were sulfon- ami es. Toward the end of the month, he was admitted to the Alton Memorial Hospital, Alton, Ill., and a tonsillectomy was performed. In May he developed right cervical lymphadenopathy that gradually increased. Five weeks before admission a generalized, pruritic, FIG.4. A, Case 4. (WU111. 56-5046). Clinical h o b graph at the peak of the reaction with the enyaqed lymph nodes appearin as generalized cervical swelling. B, Case 4. (W& Ill. 58-1109A). Occipital (wiglym h node containing pleomorphic reticulum cells 2 mitotic figures), lymphocytes, and eosinophils. (H. & E. ~ 7 5 0 . ) 174 CANCEJRanuary-February 1959 VOl. 12 FIG.5. Case 5. (WU Ill. 58-1104).Cervical lymph node showing hyperplasia of reticulum cells with eosinophils. (H.& E.~ 7 5 0 . ) maculopapular rash appeared, disappeared a costal margin. The initial impression was of few days later, and then reappeared. About 3 Hodgkin's disease. The hemoglobin level was or 4 weeks before admission, there was an in- 16.6 gm.,and there was leukocytosis of 11,750. crease in the size of the nodes, and nodes in T h e eosinophils ranged from 12% to 25% of other locations (submandibular and inguinal) the differential count. Mesantoin therapy was were visible. He was readmitted to the Alton discontinued when the patient was admitted Memorial Hospital 2 weeks before his admis- to the hospital, but Dilantin therapy was con- sion to Barnes Hospital, and a left inguinal tinued. lymph node was resected. Microscopically, the A left occipital lymph node was resected on node revealed considerable, but not total, Sept. 20, and a bone marrow aspiration was obliteration of the normal architecture. performed the following day. On microscopic , Hyperplasia of the large reticulum cells with examination, this node was similar to the first several mitotic figures was very prominent. one, except that even more eosinophils were There was a moderate amount of nuclear de- present (Fig. 4B). T h e bone marrow was bris present, but phagocytosis was not marked. thought to be normal except for an increase A great many eosinophilic leukocytes were in the number of eosinophils. Starting on seen. No Reed-Sternberg cells could be identi- Sept. 27, bromide therapy was instituted; fied. At the time of the resection, the possi- when therapeutic levels were attained, the bility that this might represent a drug reac- Dilantin he had been receiving was discon- tion was strongly considered. tinued. T h e patient was followed after his On admission to Barnes Hospital, the pa- discharge from Barnes Hospital on Oct. 20, tient showed a diffuse maculopapular ery- 1956, by the clinics of the Washington Uni- thematous rash and multiple 1- to 3-cm. firm, versity School of Medicine. A skin biopsy on nontender nodes over the occiput, in the Oct. 29 showed some lymphocytes about the cervical, submandibular and submental Iymph vessels in the upper dermis. This was inter- node groups, and in the left axilla and in- preted as nanspeufic chronic dermatitis. He guinal regions. There was a generalized swell- was started on primidione (Mysoline) on Nov. ing of the neck (Fig. 4A), and the liver was 20, 1956, and the henobarbital dosage was Bpalpated 5 fingerbreadths below the right correspondingly re uced. By February, 1957, No. 1 DRUGINDUCLEYDMPHOMA-LAIDKEENOPATHY Saltzstein 6.Ackerman 175 there was no palpable lymphadenopathy. Eosinophilsrepresented 1% to 2% of the total white blood cell count. In late September, 1957, he had some lower abdominal pain, more severe on the left side than on the right. He was seen in the clinic, and a mass of left inguinal nodes as well as an ill defined left lower quadrant abdominal mass was found. No therapy in relation to the masses was undertaken. His symptoms have since slightly abated, but the masses persist. A barium enema on Jan. 20, 1958, was inter- preted as showing a left pelvic mass. When seen again at the end of February and in early April, the inguinal mass was restricted to the region of the inguinal biopsy scar. The ill defined, nontender abdominal mass had not increased in size. cluded a hemoglobin level of 13.2 gm.,and a white blood cell count of 9,150. The differential included 13% eosinophils and 1801, monocytes. A bone marrow aspirate was thought to be normal. Old tuberculin and histoplasmin skin tests were negative. On Dec. 7, a left cervical node was resected. Grossly the node measured 2 ~ 1 x 1cm.,and was graywhite in color. The cut surface was homogeneous. On section, there was hyperplasia of the lymphoid tissue and a slight increase of fibrous connective tissue and reticulum cells (Fig. 5). No neQocris or phagocytosis was seen, and only a few scattered eosinophils were noted. Since discharge, the lymphadenopathy has completely disappeared. At the present time, the child is in perfect health. Case 5. A white boy first began having Case 6. This case has been reported pre- grand ma1 seizures at 3 years of age. On sev- viously by us.67 Briefly, it is that of a 7-year-old eral occasions a low blood sugar level was white boy who had had a convulsive disorder recorded, and at first it was thought that the since 3 years of age and also had allergic seizures were attributable to hypoglycemia. rhinitis. Three months before admission, he Later the child's mother noted that withhold- was started on ethotoin (Peganone), and 7 ing sugar, which previously had aborted the weeks before admission he developed cervical seizures if it were given in the prodromal and axillary lymphadenopathy that eventually haw, did not bring on the attacks. Phenobar- included nodes "as large as a fist." At the time htal therapy was started in August, 1956, but of admission to St. Louis Children's Hospital the patient continued to have seizures. Di- (on the service of J. Jaudon), in addition to lantin therapy was started some time after the large nodes, the liver and spleen were each that (4 capsules per day). Mesantoin therapy palpable 1 an. below the respective costal (350 mg. per day) was started on Aug. 17,1957 margins. He had a temperature ranging from (at which time the patient was already taking 37.4O to 39.2O C There were eosinophilis Dilantin and phenobarbital), and the child`s ranging from 4% to 15% in the peripheral seizures decreased in frequency. In early blood. A bone marrow examination was inter- October, 2 months prior to hospital admis- preted as being normal except for a slight in- sion, the child developed a generalized crease in eosinophils. One of the nodes was re- petechial rash that disa peared after a "shot" moved on the second hospital day, and the was given to him by [is family doctor. He Peganone therapy was discontinued at the rthen developed malaise, vague joint pains, same time. Within a matter of days, the en- anorexia, and constant low ade fever that larged nodes had all but vanished, and they continued until the time o admission. N o have not since become enlarged again. At the seizures occurred during this time, but an present time, 4 months after the resection, the episode of diarrhea occurred. This was fol- child is in excellent physical health. lowed by the "flu," and chlortetracycline Microscopically, the resected node showed (Aureomycin) was given. Six weeks before ad- virtually complete obliteration of the normal Bmission, the child develo ed right cervical architecture. It was infiltrated by a pleo- lymphadenopathy.The no es fluctuated some- morphic reaction including lymphocytes, what in size, but during the month prior to plasma cells, neutrophils, eosinophils, and the boy`s hospital admission they remained reticulum cells (Fig. 6A and B). Many of the almost constantly the same size. latter were large, with pleomorphic nuclei. On admission to St. Louis Children's Ha- pital on Dec. 3. 1957, (on the service of s. Har- and frequent mitotic figures. No Reed-Stemberg cells were found. There were many areas rison) enlarged nodes were palpated on both of necrosis with accompanying phagocytosis of sides of the neck. Some of these almost covered nuclear debris. Some invasion of the capsule the sternomastoid muscle. The nodes were by lymphocytic and reticulum elements was quite firm, and they were separate. A few seen. inguinal nodes were also plpated. The liver was felt 2 cm. below the right costal margin, Case 7. Approximately 2 or 3 months before and the rest of the physical examination was admission to Barnes Hospital, a 45-year-old within normal limits. Laboratory data in- Negro woman began having right-sided head- 176 CANCEJRanuary-February 1959 VOl. 12 FIG.6. A, Case 6. (WU Ill. 58-1456A). Pleomorphic, atypical reticulum cells (with a mitotic figure), lymphocytes, and nuclear debris in the cervical lymph node. (H. & E. ~ 7 5 0 . )B, Case 6..(WU Ill. 58-1457A). Large focus of necrosis in the cervical lymph node. (H. & E. ~ 7 5 0 . ) -No. 1 DRUG-INDUCLEYDMPHOMA-LAIKDENOPATHY Saltatein 6.Ackerman 177 FIG.7. A, Case 7. ( W U Ill. 58-2052A). Cervical lymph node showing inliltration with pleomorphic reticulum cells (with 1 mitotic figure). (H. & E. ~ 7 5 0 . )B, Case 7. jWU Ill. 58-2051A). Pericholangioliticcollections of neu- trophils, eosinophils. and lymphocytes in liver needle biopsy material. (H. & E. xS70.) 178 CANCEJRanuary-February 1959 VOl. 12 aches that tended to be worse toward the end of the day. The headaches were not accom- anied by visual auras, nausea, or vomiting, gut symptoms of hot flashes, faintness, numbness, and parasthesias, attributed to the menopause, were aggravated by them. At first, the headaches were relieved by salicylates. She was seen in the Neurology Clinic of the Washington University School of Medicine 5 weeks before her hospital admission. Combinin ahistory of headaches, which were thoug tthtoe be migraine, and the vague history of previous faintness and numbness, a tentative diagnosis of an idio athic convulsive disorder was made. B l o d studies, urinalyses, and cranial roent nograms were all normal. An electra- encepfalxam was interpreted as being "consistent wi a convulsive disorder." Therefore, the woman was started on Dilantin therapy (0.1 gm. twice a day) 1 month before admission. Three weeks before admission, she developed a generalized pruritic maculopapular rash. She was again seen in the clinic, at which time her white blood cell count was 7,000. NO differential was performed. T h e Dilantin therapy was continued. Shortly after this, she developed lymphadenopathy involving all of the palpable nodes. There was a brief episode of transient lessening of symptoms afterwards, but severe fever, chills, malaise, and swelling of her hands and feet began 4 days before her hospital admission and continued until the time of admission. She was admitted to Barnes Hospital (on the ward service) on March 4, 1958. At that time she had a temperature of 39" C. and a pulse rate of 100. She appeared acutely ill and uncomfortable. Her skin was covered with a maculopapular erythematous eruption that tended to be petechial on her legs. Large,soft, slightly tender nodes were felt in all of the node groups, but those in the left posterior cervical chain and in the axillae were especially enlarged. No abdominal organs were palpated, and no icterus was noted on examination. The white blood cell count was 8,250, with a differential count that included 5% eosinophils. The hemoglobin level was 11.3 gm.,and the hematocrit was 38%. At the time of the latter test, it was noted that the serum was icteric. Therefore, a series of "liver function" tests were performed, and they showed a total serum bilirubin level of 1.5 mg. per 100 cc. (with the direct-acting level being 0.5 mg. per 100 cc.). a serum alkaline phosphatase level of 13.1 Bodansky units, and sulfobromaphthalein (Bromsulphalein) retention of 36.2% in 45 minutes. The serum glutamicoxaloacetic transaminase was 850 units and the glutamic-pyruvic transaminase was 430 units. Serum protein levels, cephalin choles- terol flocculation, and thymol turbidity were within normal limits. A lupus erythematosus test was negative, and the erythrocyte sedimentation rate was 23 mm. per hour. Because of previous experience with reactions to hydantoin drugs, the patient was considered, at the time of hospital admission, to have such a drug reaction, and the Dilantin therapy was discontinued. Adrenal cortical hormones (prednisone) were employed in her therapy, and her headaches were relieved with salicylates and narcotics. On March 7, removal of 2 left supraclavicular lymph nodes along with skin and skeletal muscle was performed. Grossly the skin and muscle were unremarkable. One of the nodes measured 3 mm. in diameter, and the other 1 . 5 ~ 0 . 5 ~ 0 . 5 cm. Both were firm in consistency, and the architecture appeared to be preserved. There was some focal hemorrhage in the larger node. Microscopically, both nodes showed areas in which the normal nodal architecture was pre- served and areas in which it was destroyed. In the latter areas, there was infiltration of large pleomorphic reticulum cells with frequent mitotic figures, eosinophils, plasma cells, neu- trophils, and lymphocytes (Fig. 7A). No ne- crosis or phagocytosis was seen. The skin sections demonstrated a perivascular infiltration of lymphocytes, macrophages, plasma cells, and eosinophils in the upper dermis. T h e muscle sectionswere unremarkable. The nodal changes were interpreted as being those of a hydantoin-induced lymphadenopathy, while the skin changes were thought to represent a drug reaction. A liver needle biopsy was performed on March 11. Collections of inflammatory cells, especially neutrophils, eosinophils, and lymphocytes, were present about the small chol- angioles in the portal spaces and scattered elsewhere throughout the parenchyma (Fig. 7B). A few necrotic hepatic cells and other vacuolated liver cells were noted. Both bile and lipochromatic pigment were seen. No bile plugs or "bile lakes" were found. This picture was thought to be representative of a toxic cholangiolitis not unlike that described in reaction to chlorpromazine hydrochloride. Clinically, the patient improved rapidly. T h e adenopathy was virtually gone by the fourth hospital day, and the skin rash cleared up with equal rapidity. By March 19 the serum glutamic-oxaloacetic transaminase had fallen to 69 units, and the serum glutamicpyruvic transaminase to 100 units. T h e cepha h cholesterol flocculation was 3+, and the thymol turbiditv 8.5 units at this time. T h e serum alkaline phosphatase was 10.1 Bodansky No. 1 DRUGINDUCLEYDMPHOMA-LAIKDEENOPATH*Y Saltxstein Q Ackerman 179 units, and the serum bilirubin less than 0.8 mg. per 100 c c A high axillary lymph node was removed on March 21. This showed only the changes associated with dermatopathic lym hadenopathy. The patient developed an e d e m a t o u s rash postoperatively that was attributed to the pentobarbital (Nembutal) used as a preoperative medication. This cleared up rapidly when the drug was discontinued and the dosage of prednisone was increased. She was discharged from the hospital on March 30 to be followed in the clinics of the Washington University School of Medicine. Her medication at the present time includes prednisone, antacids, and vitamins. DISCUSSION The usual reactions to the anticonvulsant drugs were well summarized in the book of Goodman and Gilmans2 and that of Alexander.* Both books briefly described adenopathy resulting from Dilantin and Nirvanol therapy. The chemical structural similarities of the various hydantoin derivatives is obvious, but, as has been pointed out by Goodman and Gilmansz and others, the other anticonvulsant drugs also have similar chemical structures. This may give some basis to the similar actions and toxic effects of these drugs. It has been shown that in man and dogs, Mesantoin is demethylated to Nirvanol metabolically.1~1~2 Our third patient (Table 1, case 78) and the fourth patient (case 48) of Olmer et al.51 both showed a persistent plasmocytosis in the bone marrow and resected nodes. Neither ever showed conclusive evidence of multiple myeloma, although our patient did have abnormal serum proteins. Another case of the drug reaction mimicking appendicitis, similar to the case of our first patient (case 76), has been reported in relation to Dilantin therapy.65 This patient was not operated on, but cessation of Dilantin therapy resulted in disappearance of the symptoms. No adenopathy was described in this case. In reviewing Table 1, it can be seen that a rather characteristic clinical syndrome can be described. The lymphadenopathy most frequently involves the cervical nodes, although it can involve any palpable or visualizable groups of nodes. Perhaps this is because the cervical nodes are examined more closely than are some of the other lymph node groups, but in our fourth (case 79). fifth (case 80), sixth (case 81), and seventh (case 82) patients, the cervical adenopathy was greatly out of proportion in relation to the swelling in other node groups. T h e nodes are often tender, but they are not invariably so. They vary in size from minimal enlargement to "the size of a goose egg," as in one of the cases 4 to 8, or "as large as a fist," as in case 81. T h e swelling may begin as smn as 1week after therapy is started, or it may be delayed for as many as 18 months (case 74) or 2 years (case 13). In only 3 cases (cases 10, 13, and 74) was there a latent period of more than 4 months. T h e lymphadenopathy is usually associated with the more common manifestations of drug reactions, such as esoinophilia, fever, blood dyscrasias, and skin disorders. T h e latter include morbilliform or scarlatiniform rashes, urticaria, exfoliative dermatitis, and abnormal pigmentation. Conjunctivitis has also been reported in association with lymphadenopathy. Splenic or hepatic enlargement occurs frequently. T h e nodal involvement can occur in the absence of some of the other manifestations of a drug reaction. Only in our first case (case 76) did it occur in the absence of all of the other signs and symptoms of a drug reaction. T h e lymphadenopathy in itself is not fatal. Those patients who have died succumbed to other reactions, such as agranulocytosis, pancytopenia, or periarteritis nodosa. In the great majority of patients, the Imphadenopathy disappeared, along with the other symptoms, in 1 or 2 weeks after cessation of therapy. The gross appearance of the removed nodes is certainly nonspecific. They are enlarged, gray, and only moderately firm. Areas of hemorrhage and necrosis are occasionally seen with the naked eye. Histologically, the nodes show varying degrees of effacement of the normal architecture and a rather pleomorphic cellular reaction. Characteristically, there is hyperplasia of the reticulum cells and infiltra- tion with eosinophils. Neutrophils. plasma cells, and young lymphocytes are also seen. Mitotic fipres and abnormal reticulum cells may also be quite prominent, as in our first (case 76) and sixth (case 81) cases. Patches of necrosis and nuclear debris, the latter often present in macrophages, is quite a consistent finding. Diagnostically the problem is, as stated at the beginning of this paper, to differentiate this drug reaction from the reaction seen in the presence of the various malignant lymphomas. T h e clinical history of the patient is of major importance. T h e "drug reaction'' cause of lymphadenopathy should always be 180 CANCERJanuary-February 1959 Vol. 12 considered when a patient who has been started on a new anticonvulsant drug in the previous few months develops enlarged nodes. T h e association of the onset of adenopathy with the onset of fever and skin rash is also helpful diagnostically, but the lymphomas can also produce these reactions. A finding of eosinophilia in the peripheral blood and bone marrow would lend support to the diagnosis of a drug reaction. Leukopenia, agranulocytosis, anemia, or pancytopenia can occur as a result of either lymphomas or drug reactions. Histologically, in the drug reaction there is not the uniform picture of mature lymphocytes that there is in association with lymphosarcoma or chronic lymphatic leukemia. ReedSternberg cells are never seen, and this, along with the fact that the nodes in the drug reaction are usually discrete rather than matted, should be noted as a differentiation from Hodgkin's disease. There is also less fibrosis in the drug reaction. Otherwise, the drug reaction and Hodgkin's disease morphologically resemble each other fairly closely. The distinction between the adenopathy induced by anticonvulsant therapy and that in the acute leukemias and reticulum cell sarcoma may be more difficult. The blood and bone marrow picture is of prime importance in the former. It was only in the follow-up that we were reasonably certain that our first case (case 76) did not involve reticulum cell sarcoma. Even in the sections of this case, however, the small foci of necrosis and infiltration with eosinophils, which are characteristic of the drug reaction, were seen. This is the only case in which significant invasion of surrounding tissue was seen. Slight invasion of the nodal capsule occurred in 2 other cases of ours (cases 78 and 81). Etiologically, the adenopathy is best considered part of a patient's allergic idiosyncracy in relation to the drugs. T h e relation of the onset of the nodal swelling to the recent introduction of a new drug into the patient's therapy and the disappearance of the swelling with cessation of therapy is excellent presumptive evidence that the drug causes the reaction. In a few of the cases in this series (cases 1, 4 to 8, 17, 20, 21, 28, 29, and 78), resumption of therapy with the offending drug caused an exacerbation of the adenopathy. T h e rarity of the reaction makes the postulation of an idiosyncracy tenable. Using the figures cited by Forster,2a there must be approximately a million epileptics in the United States alone at the present time, and the majority of them must be receiving some medication. T h e search of the world's literature (Table l ) shows less than a hundred cases of lymphadenopathy resulting from therapy with the hydantoin and hydantoin-like drugs. Most authors consider the rash, fever, and bone marrow depression to be an allergic reaction. Many of the patients had a previous history of allergies. The finding of peripheral blood and bone marrow eosinophiiia and infiltration of the nodes by eosinophilic leukocytes lends more evidence to an allergic origin. Similarly, at least 3 of the cases in this series (cases 10, 74, and 75) were associated with either periarteritis nodosa or disseminated lupus erythematosus, both of which are considered by many to be allergic reactions of one type or an0ther.6~The entire course is similar to that of serum sickness,44 as was pointed out even in the days of Nirvanol therapy.68.81 Lymphadenopathy is an important component of serum sickness. Histological descriptions of the lymph nodes in serum sickness are rare.20166 GermuthSO described granulomatous and diffuse inflammatory reactions composed of reticulum cells and epithelioid cells in the mesenteric lymph nodes of rabbits given intravenous bovine serum. His photomicrographs are not too unlike our sections. The clinical and pathological similarity of the drug-induced lesions to the various malignant lymphomas leads to speculation regarding etiological relationships. Certainly, further clinical studies and animal experiments should be done with this in mind. SUMMARY A clinical and pathological syndrome closely mimicking malignant lymphomas can result from treatment with the various hydantoin and hydantoin-like drugs. This reaction should be seriously considered before any therapy directed at a lymphoma is instituted. Clinically, the syndrome includes lymphadenopathy, fever, exanthema, eosinophilia, and, less frequently, hepatosplenomegaly. Pathologically, the nodes show obliteration of the normal architecture, hyperplasia of the reticulum cells and other elements, with frequent mitoses, infiltration with eosinophilic leukocytes, focal necroses, and phagocytosis. No Reed-Sternberg cells are present. No. 1 DRUGINDUCLEYDMPHOMA-LAIKDEENOPATH*Y Saltatein 6. Ackerman 181 Cessation of therapy with the offending Etiologically, this lymphadenopathic reac- drug results in remission of the clinical and tion is felt to be the result of an allergic pathological findings. response. REFERENCES 1. AIRD,R. B.: Treatment of epilepsy with methyl- 22. FEWERMA1N..I... and VICTOROFVF.. M.: Mesanphenylethyl hydantoin (mesaatoin). California Med. toin in epilepsy:. 3 ye& study. Dis. N e w . System 10: 68: 141-146. 1948. 355-359, 1949. 2. ALEXANDEHR. ,L.: Reactions with Drug Therapy. 23. FORSIFX, F. M.: Epilepsies and convulsive dis- Philadelphia, Pa. W.B. Saunders Company. 1955; pp. orders. I n BAKERA, . B., Ed.: Clinical Neurology, vol. 165-179. 2. New York, N.Y. Paul B. Hoeber, Inc. 1955: pp. 1036- 3. A. M. A. COUNCIOLN PHARMACAYND CHEMISIRY: 1074. "Nirvanol." J. A . M . A . 99: 33-34, 1932. 24. FORSTERT, . W.: WATSONJ,. W., and NEUMARKE,.: 4. BAILEY,A. A., and WORDENR.. E.: Treatment of Agranulocytosis and thrombocyto enia following use adults for epilepsy: I, toxic effect of 3-methyl 5-5- of tridione. Lancet 1: 517-518, 194g. phenylethylhydantoin. Proc. Staff Meet. Mayo Clin. 2 4 25. FRANKEAL., L., and ROTHERMICNH., 0.:Polyar- 483-486, 1949. teritis nodosa: review together with report of case 5. BAROW, .Z.: Mesantoin (methyl-phenyl-ethyl-hy-due to hydantoin sensitization treated with cortisone. dantoin) as the anticonvulsant of choice. Am. Pract. 6. Ohio State M . J . 47: 1013-1021,1951. Digest. Treat. 5: 340-345, 1954. 26. GARVINJ,. S., and GIBES,F. A.: Mesantoin tox- 6. BELINSONL,.: Incidence of sensitivity to Mesan- icity; clinical note. Dis. N e w . System 11: 48-50, 1950. toin: effect of dosage. Dis. N e w . System 9: 252-253, 1948. 27. GAUSTADV..: Granulocytopenia and hyperkerato- 7. BERCELN, . A.; FINESILVEBR.:, SOLOMONP.,; SOMER- sis in plantae pediis after treatment with "Hydantal" FELD-ZISKINED.,, and ZISRINDE, .: Mesantoin in epi- (methyldiphenyl hydantoin). Acta med. scandinav. 127: lepsy; report of 80 cases. J . A . M . A . 143: 1460-1462, 225-232. 1947. 1950. 28. GAYLE,R. F., JR.; LYNCH,J. P., and SCHERER, 8. BEST,W.R..and PAUL,J. T.: Severe hypoplastic J. H.: Another fatality followin use of tridione and anemia following anticonvulsant medication; review of hydantoin. Virginia M . Monthly55: 288-291, 1948. literature and report of case. A m . J . Med. 8: 124-130, 29. GENTRYE... and HILL,L. F.: Granulocytopenia 1950. and death following use of trimethadione. Am. J . Dis. 9. BODARTF.,: Ober Lymphknotenschwellungen nach Child. 75: 582-586, 1948. Mesantoin-Medikation und deren cytologisches Bild. Wien.Ztschr. inn. Med. 34: 375-379, 1953. 10. BRAITHWAITRE.,F.: Fatal case of agranulocytosis associated with "tridione" therapy. Brit. M . J. 1: 14, 30. GFRMUTHF,. G., JR.: Comparative histologic and immunologic study in rabbits of induced hypersensitivity of serum sickness type. J. Exper. Med. 97: 257282. pl. 15-19, 1953. 1948. 31. GIBESF. .A.: New drugs of value in treatment of 11. BUTLER,T. C.: Metabolic demethylation of epilepsy. Ann. Znt. Med. 27: 548-554.1947. 3-methyld-ethyld-phenyl hydantoin (Mesantoin). J . 32. GOODMANL., S., and GILMANA, .: T h e Pharma- Phannacol. Q Exper. Therap. 104: 299-308, 1952. cological Basis of Therapeutics: a Textbook of Phar- 12. BUTLERT, . C.: Quantitative studies of physio- macology, Toxicology, and Therapeutics for Physicians logical disposition of S-methyl-5-ethyl-5-phenyhlydan- and Medical Students, 2d ed. New York, N.Y. The toin (Mesantoin) and 5-ethyl-5-phenyl hydantoin (Nir- Macmillan Company. 1955; pp. 178-205. vanol). J . Pharmacol. & Exper. Therap. 109 340-345, 33. GitorrER, A. L.: Diphenylhydantoin sensitivity; 1953. /13. CARNICELLTI,. ., and TEDESCCH.I,G.: Fatal acute pancytopenia fo lowing tridione treatment. New England J . Med. 238: 314-317, 1948. report of fatal case with hepatitis and exfoliative dermatitis. New England J. Med. 254 522-523, 1956. 34. HARRISONF., F.; JOHNSONR,. D., and A m , D.: Fatal aplastic anemia following use of Tridione and 14. CHAIKENB., H.: GOLDBERBC.,I., and SECALJ,. P.: hydantoin. J . A . M . A . 132: 11-13, 1946. Dilantin sensitivity: report of case of hepatitis with 35. HUNTERH, .: Treatment of epilepsy. Dis. N e w . jaundice, pyrexia and exfoliative dermatitis. New Eng- System 9: 203-209, 1948. land J . Mcd. 242: 897-898, 1950. 36. HUNTERH. .. and IENKINDS.:. Occurrence of skin 15. CHIARIH, .: Ueber das feingewebliche Bild der p i b e n t a t i o n 'during kesantoin therapy. J . A . M . A . bei Mesantoinbehandlung zu beobachtenden Lymphknotenschwellung. Wen. klin. Wchnschr. 63: 77-81, 1951. 16. COOPE,R..and BURROWRS., G. R.: Treatment of epilepsy with sodium diphenyl hydantoinate. Lancet 1: 490-492,1940. 17. DEJONC,R. N.: Current evaluation of anticonvulsant drugs. Am. J. Psychiat. 105: 767-774, 1949. 18. DEJONCR, . N.: Phenurone in treatment of psychomotor attacks. Am. J. Psychiat. 107: 825-831, 1951. 19. DREYER,R.: Erfahrungen mit Mesantoin. Deutsche med. Wchnschr. 7 9 1215-1218. 1954. 147: 744-748, 1951. 37. JONES. D. P.: Methoin in treatment of epilepsy. B d . M . J . 1: 64-67, 1951. 38. JONES, T. D., and JACOBS. J. L.: Treatment of obstinate chorea with nirvanol. J. A . M . A . 99: 18-21, 1932. 39. KASLARISE,.: Experimentelle Erzeugung von Lymphknotenschwellung durch Mesantoin. Wien. Rlin. Wchnschr. 65: 1007-1008. 1953. 40. KING,J. D.: Experimental production of gingival hyperplasia in ferrets given "epanutin" (sodium diphenyl hydantoinate). Brit. J . Exper. Path. 33: 491-498, 20. EHRICHW, .E.,and HARRIST, . N.: Formation of 1952. antibodies in popliteal lymph node in rabbits. J. Exper. 41. KOZOL,H. L.: Treatment of epilepsy with Med. 7 6 335-348, pl. 26-27, 1942. methyl-phenyl-ethyl hydantoin (mesantoin). Res. N m . 21. FETIXRMA1.NL... and VICIVROFFV.. M.: Practi- Q Ment. Dis. Proc. 26: 404-411, 1947. cal aspects of Meiantoin therapy in epilepsy. A m . J . 42. LEARDS,. E.: GREERW, .E. R., and KAUFMAN, Psychiat. 105: 410-416, 1948. I. C.: Hepatitis, exfoliative dermatitis and abnormal 182 CANCERJanuary-February 1959 VOl. 12 bone marrow occurring during tridione therapy; report of case with recovery. New England J. Mcd. 240: 962-966, 1949. 43. LINDQVISTT, .: Lupus erythematosus disseminatus after administration of mesantoin. Acta med. scandinav. 158: 131-138, 1957. 44. LONGCOPEW, . T., and WINKENWERDEWR. , L.: Anaphylaxis, serum disease, urticaria, and angioneurotic edema. In HERRICKW. . W., Ed.-in-Chief: Nelson New Loose-Leaf Medicine, vol. 2. New York. N.Y. Thomas Nelson & Sons. [1941]; pp. 631-648. 45. LOSCNZO,A. E.: Mesantoin in control of epilepsy. Neurology 2: 403-411, 1952. 46. MARTIN,R.; SUREAUB, .: VERON,M., and BARME,M.: Accidents cutan& e t ganglionnaires au cows d'un traitement par la diphkny1-hydanto.he et la phknadtylurk. Arch. franc. pCdiut. 11: 979-982, 1954. 47. MARTINEZP,. D., and GOMEZOROZCOL,.: Scarlatinform dermatitis produced by administration of mesantoin. Bol. mCd. Hosp. inf. 7: 884-886, 1950; abstr. in A. M . A . Am. J . Dis. Child. 84: 286, 1952. 48. MELLERR, . L.,and RESCH,J. A.: Convulsive disorders in adults. Postgrad. Med. 6: 452-458, 1949. 49. M E R R ~H, . H., and PUTNAMT, . J.: Sodium diphenyl hydantoinate in treatment of convulsive disorders. J.A. M . A . 111: 1068-1073, 1938. 50. MONFORTJ,. A.: Chorea treated with phenylethyl-hydantoin and typhoid vaccine; a comparative study. Am. J. Dis. Child. 47: 1269-1278. 1934. 51. OLMERJ,.; PAILLASJ, .; ROGERI,.; MURATORRE.,, and BADIERM, .: Manifestations ganglionnaires au cours de traitements par la m&hyl-3-phbnyl ethyl-5-5-h dan tohe. Bull. et men. SOC. m i d . H6p. Paris 68: 838-8411 1952. Also in Presse m t d . 6 0 1748-1750, 1952. 52. POYNTONF., J., and SCHLESINGEBR.:. Treatment of chorea by nirvanol. Lancet 2: 267-270, 1929. 53. RAY,H. H., and CUNNINGHAJM. ,S.: Phenylethylhydantoin in treatment of Sydenham's chorea. A m . J. Dis. Child. 3 9 1205-1213, 1930. 54. RICH, A. R.: Hypersensitivity in disease; with especial reference to periarteritis nodosa, rheumatic fever, disseminated lupus erythematosus and rheumatoid arthritis. Ha- Lcct. 42: 106-147, 1946. 55. RINCERTZN, ., and ADAMSON.C. A.: Lymph-node response to various antigens; experimental-morphological study. Acta path. et microbiol. scandinav. suppl. 86: 1-69, 1950. 56. ROHRBACHP,.: Knochenmarksschadigung und Panhbocytopenie durch Hydantoinkorper. Schweiz. med. Wchnschr. 80: 337-340, 1950. 57. SALTZSTEINS.. L.; JAUDONJ., C.; I.USE. S. A., and ACKERMALN., V.: Lymphadenopathy induced by etho- toin (Peganone); clinical and pathological mimicking of malignant lymphoma. J . A . M . A . 167: 1618-1620, 1958. 58. SCHICKB,.; SOBOTKAH,., and PECK,S.: Chemical allergy and nirvanol sickness. A m . J. Dis. Child. 45: 1216-1220, 1933. 59. SCHLESINGEBR.;, POYNTONF,. J., and EAST,T.: Discussion on use of nirvanol in treatment of chorea. Proc. Roy.SOC.Med. 23: 468-472, 1930. 60. SCHWABR,. S.; TIMBERLAKWE,. H.,and ABBO-IT, J. A.: Control of side elfects of anticonvulsant drugs. M . CIin. North America 38: 1339-1350, 1954. 61. SILBERI,. B., and EPSTEINJ,. W.: Treatment of chorea with phenyl-ethyl-hydantoin; study of 28 cases. Arch. Pediat. 51: 373-382, 1934. 62. STEINBERSG.,H.: Reaction to diphenyl hydantoin sodium; report of case simulating Rocky Mountain spotted fever and review of literature. M . Ann. District of Columbia 2 2 600-603, 1953. 63. TISDALL,0.R.: Nirvanol treatment of chorea. Arch. Dis. Child. 5: 397-404, 1930. 64. VANWYK.J. J.. and HOFFMANNC.. R.: Periarteritis nodosa. Arch. Int. Mcd. 81: 605-611. 1948. 65. WEAVER0,.M.:HARRELLD,. L., JR., and ARNOLD, G. B.: Report on use of sodium 5, 5-diphenyl hydantoinate in 14 selected cases of epilepsy at Virginia State Colony for Epile tics and Feeb!e-Minded. Virginia M . Monthly 66: 52!-527, 1939. 66. WISE,R. A., and HEUMANWN,. A.: Mesantoin in treatment of epilepsy. Psychiat. Quart. 29: 403-411, 1955.