POLYARTERITIS IN CHILDHOODBY

A. V. NEALE, M.D., F.RC.P.(From the Department of Child Health, University of Bristol)

With increased study of the mesenchymal tissuesand their special reactions to toxic, infective, andallergic factors, the parficular pathological grouping,in regrd to the place in which the blood vessels aredirctly involved, has been widely considered. Aconsiderable literature daling with the pathologyand the clinical miftations of polyarteritis hasshown clearly that the connective tissue eleents inthe vessel walls are, under certain conditions, highlysusceptible to inflammatory and necrotizing agents.It must be admitted that the morbid anatomy hasbeen clearly -defined, and that cetain clnial andexperimental observations have lead to some supportfor an etiological basis of allergy. Neverthekss theprecise relationships cannot be considered to befully understood. In fact it is quite evident that inthis field of pathology there is still ample scope forclassiation, and particuLaly so in as much as asimilar tissue response nism is concerneda wide variety of clinical forms of disease. Poly-artertis may in some cases exist as the only apparenttissue manifestation, but in others there is a morewidespread nymal tissue reaction, and aninfinity of graduations may occur.Teilum (1946), in discsing lupus erythematosus

disemninatus and related di , notes the peculiarfibrinoid necrotizing proceses in free connectivetissue and in the walls of blood vessels together withmiliary epithelioid cell granulomata in the serosa;he proceeds to consider 'a state of allergy inducedby specific processes of immnunity ' as a basis for theissue chang. In contrast to lupus erythematosusdisinatues Teilum considers rmatic fever tobe etiologically and cinially a tye of specificlesion. However, this author is mindful of theclose pathological relationship of lupus eythe-matosus disinatus, polyarteritis and arteriolotisgranulomatosa (allrgica), and inlines to the viewthat there is a group of ' para-rheumatic ' diseaseof simila pathognesis and differt etiology, butposse g in common a state of hypersensitivity tobacterl toxins and allThe cinical and experimental observation of Rich

(1942) has ehluidated possible etiologal factorsinvolved in polyatritis nodosa. There is in theexperiments ample evidenc for Rich's confidentbelef that arterialesons of this type are mnifesta-tions of anaphylactic snsiivity. Undoubtedly

there are widely different types of sensitizingantigens which have the potentiality of causingpolyartertis in man. Galan (1945), after experi-mental evidence, is of the sme opinion. There is,however, the necessary adjuvant factor of con-stitutional and special tissue susceptibility whichtends to defy clinical recognition.

Linkage in pathology are revealed in the case ofrheumatic carditis associated with characteristicpolyarteritis nodosa (Neale and Whitfield, 1934).Friedberg and Gross (1934) also describe four casesof periarteritis nodosa in rheumatic fever in whichAschoff bodies were demonstrable in the myo-cardiumn Buckley (1946) points out that in theLibman-Sacks syndrome vascular chang idenicalwith those seen in polyarteritis nodosa are some-times present, and he records an example in whichgeneralized vasculitis, in association with ulcerousdennal changes, were present; he considered thewhole pathology to be a sequel to chronic focalinfection.The experimental work of Selye (1946) is remark-

able in so far as it reveals, in certain animal, anarterial reaction with nodular inflammatory changesfollowing injections of adrenal cortical steroidhormone or anterior pituitary extract; this effectis inteniie if the animal is receiving excessiveamounts of salt. Selye advances the view that insome susceptible persons an overdosage of endo-genous cortical hormone, such as may occur inreaction to a sore throat, a nervous shock, or anexposure to cold, etc., may detemine the occurrenceof tissue reponses istngishable from that ofrheumatic fever and periarteritis nodosa.

It is signiicant that eosinophilia is only exception-ally present in childWood cases of polyarteritisnodosa, and this bears no inhlert relationship tothe degree, extent, distribution, or the severity ofthe lesions. In Wilmer's (1945) two very severeinfant cases there was no increase in eosinophils.In a series of twenty-eight cases of pobarteritis inchildren, two only also had asthma, whereas ofthree hundred at all a with the same arterialdiseas, fifty-four had asthma and 94 per cent. ofthese revealed hyper-eosinophilia (Wilson, 1945). Inthe absence of associated asthma pobarteritis issi cantly dis ted from eosmnophilic response.Although the resech of Rich and others

have placed a definite alergic esponsibilit inpolbaitis, clinical observations are by no meansso cear, possily owing to in di es n

224

by copyright. on June 16, 2021 by guest. P

rotectedhttp://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.24.119.224 on 1 S

eptember 1949. D

ownloaded from

http://adc.bmj.com/

POLYARTERITIS IN CHILDHOODassessing prodromal phenomena in the disease orto the virtual impossibility of exploring sufficiently,at a clinical level, allergic mechanisms of this order.It is of some interest that polyarteritis in childhoodis rarely associated with any known common formor type of allergy. This suggests that in poly-arteritis a specially individualized arterial receptormechanism exists for allergen action. By and large,the incidence of this unusual vascular susceptibility,at any age group, appears to be considerably lessthan the clinical allergies of other tissue systems.From the clinical aspect no age group appears to

be immune in regard to arterial susceptibility andthe condition is probably not of great rarity ininfancy. The bizarre nature of the clinical manifes-tations may well be confusing, but Wilmer (1945)has described two cases of polyarteritis in the firstmonth of life. In each of these two infants sepsiswas present and the nature of the violent and ful-minating course of the disease would indicate aprobable peculiarly inherent arterial tissue sensi-tivity. The intensity of this susceptibility is revealedin the case of a baby aged nine months who diedsuddenly two months after an attack of acutetonsillitis: the coronary arteries were severelyaffected with nodular arteritis (Diaz-Rivera andMiller, 1946).Extended search will probably reveal that poly-

arteritis is more frequently present than is usuallyconsidered. Moreover, there is a strong tendencyto some special sectional distribution of the arterialinflammatory reaction in different patients or theeffects are dominantly represented in a particularvisceral, neural, or cutaneous zone. The prognosisfrequently depends upon the more particularlyindividualized locality of the disease and, in fact,clinical studies suggest that focal arterial reactions,e.g. cerebral, renal, coronary, may be more devastat-ing than quite extensive cutaneous involvement.This tendency to a local dominance is well illustratedby Malamud (1945), who describes a child agedfive years in whom polyarteritis selectively involvedthe meningeal vessels with resultant cerebral damageand decerebrate rigidity. Ophthalmoscopy may bespecially helpful; inspection of the choroidalarteries may show fusiform aneurismal dilatations(Goldsmith, 1946) or degrees of retinal detachment(Sampson, 1945).Although visceral involvement is rightly regarded

as a constant menace in polyarteritis, and Bradley(1947) considers the outstanding and special dangerto the renal vascular mechanism, cutaneous lesionsmay exist as the only apparent pathological distribu-tion. Downing (1947) believes that the cutaneousarteries are associately affected in 40 per cent. of allcases, but it has no special dominance in any agegroup.On clinical and pathological evidence the disease

may have an exclusively subcutaneous localization.Miescher (1946) described four examples of thiskind, and in one form the leucocytoclastic vasculitisof the subpapillary vessels is likened to the papulo-

haemorrhagic exanthema of the type peliosisrheumatica. In another, the appearances weresimilar to that of a papulonecrotic tuberculide, andyet in the others there were variable histologicalappearances, sometimes approaching granulo-matous changes.The cutaneous type of polyarteritis nodosa may

reveal severely acute and alarming effects, moreespecially owing to the possibility of ischaemicnecrosis. Veran (1945) observed massive gangreneof the foot as a result of the disease in a baby agedfifteen months, but in whom general recoveryoccurred after several months' illness. Particularlymutilating effects were observed by Galan in a boywho developed gross necrosis of the abdominalwall and gangrene of the terminal phalanges of thefingers. In this case there was also a ilivedoreticularis ' and a phasic variation of 0-10 per cent.in the eosinophil polynuclear leucocytes. Despitethe severe cutaneous effects, slow but satisfactoryrecovery occurred. Peripheral panarteritis followingscarlet fever is recorded by Hoyne and Smoller(1941).An appraisal of the clinical signs of the disease

in childhood (44 cases) revealed purpuric rash in34 per cent., other rashes, including urticarial type,in 27 per cent., and palpable nodules in 4- 5 per cent.(Keith and Baggenstoss, 1941).The cutaneous form of polyarteritis offers special

advantages in the clinical study of the disease.

Case ReportsCase 1. D.J.W. was a girl aged nine years, the

first child of two. There was no family history ofdisease. The child's past history was one ofpertussis and measles only. In May, 1943, thepresent illness began. The child developed a fairlysevere acute streptococcal tonsillitis with moderatesubmaxillary adenitis. A few days later' rheumatic'pains commenced in the shoulders, back and legs,and shortly afterwards an ' unusual ' skin rash madeits appearance. The child felt generally ill andremained in bed until admitted to hospital. Underhospital observation it was clear that a somewhatunusual syndrome was present. She was irritableand in continued discomfort and had variable limbpains. The two outstanding visible signs were thepersistent flexural position of the limbs in bed anda very marked mottled rash all over the body,especially marked on the limbs. The rash wassimilar to that of 'cadaveric staining ', and wasstriking in so far as the face was also affected andthe intensity varied from day to day, sometimesbeing peculiarly prominent and unlike any otherknown rash. There was a fluctuating appearanceand disappearance of subcutaneous tender noduleswhich left no doubt that the subcutaneous arterieswere affected by acute focal, nodular, and inflam-matory changes. There was never, in this girl, anyevidence of tissue or skin necrosis, which presumablyindicated that the arterial lumen was in no area

225by copyright.

on June 16, 2021 by guest. Protected

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.24.119.224 on 1 Septem

ber 1949. Dow

nloaded from

http://adc.bmj.com/

!PCHIL-E.( PL.ETISLE I CHILDHo)OP

-

I4

t

5.

*jKi'.

e.S

n~ F - -D.- A --- ; - ^s -s- ->-- -_ .S E --. - v \.. . . !,, l.. _ _ . _ t_ _ ...... x .. _ _ . ... .. .. . ... . W . _ _ ... . _

.. -_D_ -' _- -.

F: -E)! \\.T . i-_ !

- .: - \ ,>

.X. s \_ _ S _ _\ . - . . _ . ... : ,:_F . . _ kO- N _%

.rfl2_ 0

C F: * -D 7 \. B: @s - s--s-P-t : : -" . . _ . _ , C . _ . . s vE h_ ai i Ii _1 ' ' > --s . _ _ .. ., . s . . s ... t . x- _ v e e e _e * .... . C:> ;.*v t . _ .. W. t ! > t . >s. * _>w v \ _% h; ., b _ v k . W . _ _ _ B e . . s s _; b h1 --t tu :is- s E:_>. - ..... _s -.- ;: : . x e . . . s _ v s . v : >: _ : z : C: .F , J".VF_Illlw. ...,rL,"., - -1-4hA . by copyright. on June 16, 2021 by guest. Protectedhttp://adc.bmj.com/

Arch D

is Child: first published as 10.1136/adc.24.119.224 on 1 S

eptember 1949. D

ownloaded from

http://adc.bmj.com/

POLYARTERITIS IN CHILDHOOD 227by copyright.

on June 16, 2021 by guest. Protected

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.24.119.224 on 1 Septem

ber 1949. Dow

nloaded from

http://adc.bmj.com/

ARCHIVES OF DISEASE IN CHILDHOODobliterated. There was no evidence at any time ofany arterial disease in the deeper structures or in theviscera. Repeated examination revealed no evidenceof renal involvement and the heart remainedclinically and electrocardiographically normal. Theoutstanding clinical feature was the persistent rash,traces of which can still be seen (three years later).Considerable general wasting accompanied theillness, and over an observed course of eight months(May, 1943, to February, 1944) there was anundulating pyrexia with maximal levels around 1030.The pulse rarely reached above 120 and therespiratory rate was 30. There was no abnormalsweating. There was a persistnt lucocytosis of15,000 per c.mm., but eosinophils were never morethan 4 per cent. Blood cultures were sterile.A biopsy of the skin (forearm) was done on

March 14, 1944. The majority of the arteries in thesubcutaneous tissue, and in a lesser degree thoseof the cutis vera, showed severe pathologicalchanges. There was proliferation of the sub-endothelial connective tissues of the intima,necrobiotic changes of the media, and proliferationof the vascular granulation tissues of the media withmuch disarrangement of the muscle fibres. Thecellular infiltrations consisted of neutrophil poly-morphs, lymphocytes, monocytes, and a feweosinophils. The infiltration extended sometimesthroughout the whole vessel wall and often diffruselyinto the perivascular connective tissue. Foci ofextensive necrosis were seen in the perivasular tissue,and a form of giant cell granulation tisse could beseen in some areas. There was occasional dis-ruption of the muscle fibres. The pictue was thatof panarteritis nodosa.From time to time the child had severe limb pains

and very strongly resented being handled or moved.Skin nodules appeared and disappeared, and thewho illness presented considerable phasic vari-ability. Many symptomatic remedie were pre-scrbed, but it seemed clear that the diseas wasrefractory. In view of the possibility of retainedtonsil sepsis, tonsillectomy was performed, but noappreciable improvement occurred. Three monthslater a very severe exacerbation of all the symptomsand signs appeared with some oedema of the feet.The mesh-like general rash and erythematousnodular foci again became prominent. The clinialprogress indicated a waxing and waning of arteritisnodosa of cutaneous form.

After a prolonged stay in hospital, gradualimprovement occurred, although limb pains occurredwhen she was unduly warm in bed. The rashgradually diminished.

In the following two years occasional crops ofperiarterial nodules made their appearance and wereusually accompanied by limb pains and some daysof immobility. Finally, in 1945, she semed to befree from all effects of the prolonged illess. Oneyear later she developed diphtheria and was givenadequate doses of anti-serum without any adverseeffects. In 1948, at the age of thirteen years, she

was cliicaly well. Puberty occurred normally andmenstruation appeared.

Case 2. H.J. was a boy aged eight years and thefifth child of seven. There was no family historyof disease, and the child had a past history ofmeasles only.

This boy was admitted to hospital in June, 1943,seventeen days after a sore throat, and actuallyduring the same time as the previous case. Theclinical picture at that time was identical. Thirteendays before admission he had received sulphonamideand three days later a Macular rash appeared allover the body and also numerous red-purple focalswellings. Pains in the joints, varying in the mannerof rheumatic fever, also occurred. There was noevidence of any visceral disturbance.

Histology of a subcutaneous nodule confirmedthe diagnosis of periarteritis nodosa. The bloodshowed a fluctuating neutrophil leucocytosis,reaching a maximum of 21,400, but at no time didthe eosinophils exceed I 5 per cent.

This boy was particularly troubled with limbpains, and on several occasions temporary jointswellings with small effusions appeared. Clinicallythe intensity of the disease was more severe in thiscase. The ' mesh-work ' erythema was very markedand quite prominent on the face. The feet wereswollen and very painful and there were manysubcutaneous nodules along the course of thearteries. The heart was clinically normal andelectrocardiograms were normal. A variablepyrexia with maxima of 103° continued over manyweeks. Intensive doses of penicilln had no favour-able effect, and sulphathiazole was given without anyresponse. Over a period of one year the diseasevaried in intensity, but eventually he becameambulatory and comfortable, and his motherthought he was very well. However, a few cutaneousarterial nodules continued to appear at intervals andthe ' cadaveric ' staining on the limbs was persistent.Eighteen months after the onset the boy was gainingweight and had no symptoms, but he had clearlynot reached a phase of real health. Three weekslater he had a 'cold ' and a recurrence of theprevious limb pains and a marked accentuation ofthe cutaneous erythema. His condition quicklydeteriorated and necrosis appeared in the fingersand in some other skin areas. Necropsy revealedno evidence of any visceral involvement in thearterial disease. There was no tonsil sepsis or anyother evidence of active focal infection. There wasno renal pathology.

DisassThese two cases of polyarteritis indicate (1) that

the disease reaction may be apparently confined toa particular section of the arterial system; (2) thatsuch pathological response may have a prolongedand clinically fluctuant course and be compatiblewith a satisfactory recovery without sequelae; or,

228by copyright.

on June 16, 2021 by guest. Protected

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.24.119.224 on 1 Septem

ber 1949. Dow

nloaded from

http://adc.bmj.com/

POLYARTERITIS IN CHILDHOOD 229(3) by an acute exacerbation, lead to quicklyspreading ischaemic necrotic changes.The clinical features of the dermal form of

arteritis are distinguishable more particularly by theoccurrence of painful nodular swellings visible andpalpable in the skin. The arterial nodules may beextremely numerous and yet apparently the incidenceof intra-arterial thrombosis is minimal. This isparticularly indicated in the case of D.J.W., in whomit was not possible at any time during her relapsingillness of many months to detect any effects arisingout of lumenal occlusion. The type of generalizedrash is peculiarly characteristic in its meshwork ofbluish discoloration and resemblance to that seenin cadaveric staining; this is most marked on thelower limbs and is very persistent throughout, andfor some time after the active phase of the diseasehas subsided. In the two cases here recorded theface showed a variable general swelling in relation-ship to the periodic intensification of the rash.The mucous membranes were unaffected butconjunctival suffusion was present.No definite etiological factors were discovered,

but in each case the arteritis was preceded by rathersevere sore throat, so that bacterial toxaemia wasprobably causally related. In neither case could anydefinable allergic susceptibility be established in anindividual or familial basis. The bouts of fevermay be merely related to pyrogenic products in thenecrotizing arteritis.

TreatmentThe life history of the disease and its refractory

nature to any known specific therapeutic measurereduces the treatment to symptomatic remedies forpain, insomnia, etc. The tendency for lower limbcontractures, owing to prolonged decubitus immobi-lization, must be overcome by suitable temporarysplinting. The peripheral vascular ischaemic effectsmight be in part remedied by nicotinic acid.Benadryl and other anti-histamine drugs have alsoreceived clinical trial, without, however, anydetectable effect.

aummrySome considerations of the possible etiological

relationships in polyarteritis in childhood arepresented and two cases of the 'dermal ' form arerecorded. A prolonged illnes, with an ever-presentmenace of possible visceral involvement, wassuccessfully negotiated and clinical recovery withoutsequelae occurred in one child. In another child theprolonged and variable clinical course seemed tohave reached a favourable outcome, but unexpect-edly a sudden and intense relapse quickly progressedto acute obliterative cutaneous arteritis and spread-ing ischaemic necrosis.

REFERENCESBogeart, L. van et al., (1932). Ann. Med., 31, 530.Bradley, E. J. (1947). J. Pediat., 31, 78.Buckley, S. (1946). Helvet. Paediat. Acta., 1, 524.Downing, J. G. (1947). New Eng. J. Med., 237, 906.Diaz-Rivera, N. S., and Miller, A. J. (1946). Annals

Int. Med., 24, 420.Friedberg, C. K., and Gross, L. (1934). Arch. Int. Med.,

54, 170.Galan, E. (1945). Bol. Soc. Cubana Pediat., 17, 293.Goldsmith, J. (1946). Amer. J. Ophthal., 29, 435.Hoyne, A. L., and Smollar, L. (1941). J. Pediat., 18,

242.Keith, H. M., and Baggenstoss, A. H. (1941). J. Pediat.,

18,494.Malamud, N. (1945). J. Neuropath. exp. Neurol., 4, 88.Middleton, W. S., and McCarter, J. C. (1935). J. Amer.

Med. Sc., 190, 291.Miescher, G. (1946). Dermatologica Basel, 92, 225.Neale, A. V., and Whitfield, G. W. (1934). Brit. med. J.,

2,104.Rich, A. R. (1942). Bull. Johns Hopk. Hosp., 71, 123.

and Gregor, J. E. (1943). Ibid., 72, 65.(1945). Ibid., 77, 43.

Sampson, R. (1945). Brit. J. Ophthal., 29, 282.Selye, H. (1946). J. clin. Endocrinal, 6, 117.Teilum, G. (1946). Acta. med. scand., 123, 126.Wilmer, H. A. (1945). Bull. Johns Hopk. Hosp., 77, 275.Wilson, K. S., and Alexander, H. L. (1945). J. Lab.

clin. Med., 30, 195.Veran, P. (1945). Arch. Mal. Coeur., 38, 149.

by copyright. on June 16, 2021 by guest. P

rotectedhttp://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.24.119.224 on 1 S

eptember 1949. D

ownloaded from

http://adc.bmj.com/