The Surgical Implications of Purpura Fulminans

The Surgical Implications of Purpura Fulminans Jon R. Cohen, MD, Rudy Lackner, MD, Alex Keller, MD, Barry Douglas, MD, N e w H y d e Park, N e w York, and Stony Brook, N e w York

Purpura fulminans is an uncommon catastrophic syndrome that occurs in children, typically one to four weeks after a seemingly benign infectious process. The child usually presents with a high fever, purpuric ecchymosis, hypotension, disseminated intravascular coagulation, and gangrene of the extremities. We have recently treated six children, whose mean age was 22 months; three were male and three were female. Five of the six had a change of mental status upon initial examination. Their mean temperature was 104° F. All six children had purpuric involvement of their extremities; three had involvement of their hands, two had involvement of their faces, and two had involvement of their trunks. All had absent palpable pulses and sluggish capillary refill in the involved hands and feet. Two patients died shortly after admission as a result of severe end-stage sepsis. The platelet counts in these two patients, and the white blood cell counts were markedly depressed. The mean platelet count of the survivors was 370,000 and the mean white blood cell count was 25,000. Lumbar punctures were positive for bacterial meningitis in five patients and viral meningitis in one patient. All patients were treated with intravenous heparin. Of the four survivors, two lost significant tissue and required multiple plastic reconstructive procedures, and two improved on heparin alone with no tissue loss. In addition to systemic support and intravenous antibiotics, the mainstay of treatment is one of immediate heparinization and a continuous heparin drip. Heparin prevents subsequent small vessel thrombosis and limits tissue loss due to ongoing purpura. Conservative management of the purpuric lesions is the treatment of choice until final demarcation occurs. (Ann Vasc Surg 1990;4:276-279) KEY WORDS: Purpura fulminans; infection; heparin therapy; disseminated intravascular coagulation.

Purpura fulminans is a necrotizing purpuric ecchymosis that typically occurs in children one to four weeks after a seemingly benign infectious process [1-3]. It typically follows such childhood Presented at the Annual Meeting of the Peripheral Vascular Surgery Society, New York, New York, June 17, 1989. From the Department of Surgery, Divisions of Vascular and Plastic Surgery, Long Island Jewish Medical Center, and Schneider Children's Hoapital, New Hyde Park, New York, and the State University of New York. Stony Brook, New York. Reprint requests: Jon R. Cohen, MD, Department of Surgery, Long Island Jewish Medical Center, New Ityde Park, New York 11042.

diseases as varicella, scarlet fever, rubella, meningococcemia, measles, or any typical streptococcal infection [4]. The clinical course is typically one of high fever, purpuric ecchymosis of the extremities, hypotension, disseminated intravascular coagulation (DIC) and frank gangrene of the upper and lower extremities. We recently treated six such children with purpura fulminans, the data for which form the basis of this report.

CLINICAL MATERIALS The records of all children with purpura fulminans admitted to the Schneider Children's Hospital 276

S U R G I C A L I M P L I C A T I O N S OF P U R P U R A F U L M I N A N S

VOLUME 4 N o 3 - 1990

277

TABLE I.wClinical data on six patients Age 4 yr

Sex M

2

21/2 yr

F

Lethargy

Diffuse rash DP/ PT absent 105 °

Extent of purpura Gangrene of toes multiple areas, face, extremities Gangrene on fingers of both hands

3

91/2 mo

M

Cough Rhinorrhea

Nuchal rigidity 102 °

Purpura, both feet

4

11 mo

F

Cough Rhinorrhea Irritability

Nuchal rigidity 106 °

Gangrene on fingers of left hand

5

25 mo

F

Lethargy

104 ° Nuchal rigidity

Purpuric lesions, hands and trunk

102 o Rash Extremity swelling, Nuchal rigidity

Purpura on trunk, extremities, face

Patient 1

6

8 mo

M

Signs Symptoms DP/PT absent Lethargy Nuchal rigidity Photophobia 105 °

Irritability

over the last five years were reviewed. Charts and records were reviewed and analyzed for age, symptoms, signs, extent of purpura, laboratory values, treatment and outcome. Since 1984, six children with purpura fulminans were identified. Clinical data on these six are presented in Tables I and II. The mean age of the six patients was 22 months. There were three boys and three girls. Five of the six patients had a change in mental status on initial examination. Three of the six children were lethar-

TABLE II.wSummary of clinical data Mean age Sex M-F Change in mental status Lethargy Irritability Mean Temperature Mean white blood cell count Survivors Deceased Mean Platelets Survivors Deceased Meningitis Bacterial Viral Tissue Loss in Survivors

22 months 3-3 5/6 3/6 2/6 104 ° 25,000 2.2 308,000 80,000 6/6 5/6 1/6 2/4

Labs WBC 49,000 Plat 793 K PT 10.8/11.2 PT 27.4 FIB 340

LP 40 WBC 96% Poty Prot 29 Glu 149 Meningococcus

WBC 19,000 Plat 60 K PT 15.4/10.7 FIB 257

81 WBC 31% Poly(27 L) Prot 24 Glu 79/55 (+) Group C Meningococcus 5 WBC 1 poly (88 L) (75% Atypical)

WBC 20.4 PET 330 PT 10.5/10.2 PTT 37 WBC 3.0 PT 16.8/10.9 PTT > 100 Fibrinogen 146 Platelets 70 WBC 1.5 PT 20/15 PTT 30 Fibrinogen 210 Platelets 90 WBC 16.6 PT 9.5/10.5 P'IT 33 Fibrinogen 250 Platelets 90

(+) Meningococcus

(+) Pneumonococcus

(+) Pneumonococcus

gic and two of the six irritable. Five of the six patients had nuchal rigidity on examination. The mean temperature of the six patients on admission was 104° F. All patients had purpuric involvement of their extremities, three had involvement of their hands, two had involvement of their face, and two had involvement of their trunk. Figure 1 shows a patient with gangrenous extremities. The white blood cell (WBC) count was variable depending on their degree of sepsis at the time of presentation. In the two patients that died shortly after admission, the WBC counts were abnormally low (1.5 and 3.0), reflecting the severity of their sepsis. Platelet counts, like WBC counts, were a reflection of the ongoing sepsis. In both of the patients that died platelet counts were 90,000 and 70,000. The mean platelet count for the survivors was 370,000. Fibrinogen values were normal in five of the six patients. All six patients had positive lumbar punctures. Five of the six patients had bacterial meningitis and one had viral meningitis. The two patients that died had purpuric lesions at the time of their death but did not survive long enough for us to determine whether or not they would have developed tissue loss. Of the four survivors, two lost significant tissue and required plastic reconstructive procedures while the other

278

S U R G I C A L I M P L I C A T I O N S OF P U R P U R A F U L M I N A N S

ANNALS OF VASCULAR SURGERY

~ - ~ f L~j

Fig. 1. Typical appearance and distribution of purpuric lesions.

two improved on heparin alone, thus avoiding subsequent tissue loss. None of the patients had palpable pedal pulses initially.

DISCUSSION Purpura fulminans, left untreated, is uniformly fatal with multisystem failure including respiratory, renal, and finally cardiac failure. The benign initial presentation was impressive, as was the rapidity with which all of these children deteriorated. It is worrisome that even in retrospect, no specific feature of the initial clinical presentation stands out as an indicator that might help to identify the disease at an earlier stage. With the large number of children who contract upper respiratory tract infections, colds, flu, and otitis media, it may be impossible for primary care physicians to select those children at risk for developing purpura fulminans. An interesting feature is the altered coagulation profile consistent with disseminated intravascular coagulation. Antley and colleagues reported depressed levels of factors I, II, V, VII, VIII, and IX m addition to thrombocytopenia in patients with purpura fulminans [5]. Interestingly, our six patients showed variable alterations in their blood coagulation status, which seemed dependent on the degree of sepsis. We postulate that since all four survivors had heparin administered before disseminated intravascular coagulation became clinically evident, more severe alterations in their coagulation may have been averted (Table I). The underlying pathological process is one of perivascular infiltration by inflammatory cells and intravascular thrombosis. Histologically, there is extensive subcutaneous necrosis, perivasculitis, and small vessel thrombosis [6]. The end result is a specific injury to the small vessels that produces

Fig. 2. Mechanisms and manifestations of purpura fuIminans. (a) Lower extremity with purpuric lesion. (b) Microscopic section of skin and muscle biopsied from purpuric lesion. Note intravascular coagulation of musculocutaneous perforators. (c) Detail of thrombosed musculocutaneous perforators. (d) Distribution of lesions on hand. (e) Distribution of lesions on foot. (f) Distribution of lesions on leg and knee.

capillary thrombosis, extravasation of blood, and patchy necrosis [7] (Fig. 2). Most authors agree that the etiology of the coagulation defect associated with purpura fulminans is tied to the effects of disseminated intravascular coagulation [8,9], similar to the coagulation defect of the Schwartzman reaction in rabbits [5,10--12]. In an extensive review of purpura fulminans cases reported since 1960, Chu [4] listed the known etiologic agents associated with the disease. Table III lists the etiologic agents reported since 1960 in addition to our six cases. Interestingly, all of our patients had meningitis including bacterial meningitis in five and viral meningitis in one (Table II). In addition to specific system support and intravenous antibiotics, the mainstay of initial treatment from a consulting surgeon's standpoint is one of immediate full heparinization and a continuous heparin drip. Of particular clinical importance is the treatment of all coagulation defects with heparin therapy to reverse the pathological process. Heparin reverses the factor abnormalities and prevents subsequent

VOLUME 4 NO 3 - 1990

SURGICAL IMPLICATIONS OF PURPURA F U L M I N A N S

TABLE tll.--Etiologic agents in 73 patients [4] No. of patients Meningococcus Pneumonocoeeus Staphylococcus Streptococcus E. coil Neisseria Klebsiella Proteus Varicella Measles Viral meningitis Snake, d o g bites Vibrio p a r a h e m o l y t i c u s Rickettsia Leptospira Undetermined

15 7 6 4 2 1 1 1 7 4 1 2 1 1 1 20

Mortality 3 3 4 2 2 -1 1 m 1 -----

279

current m a n a g e m e n t includes c o n s e r v a t i v e w o u n d care until final d e m a r c a t i o n occurs. This usually consists of dry sterile dressings, h o w e v e r , on occasion dry gangrene can convert to wet purulent tissue requiring early debridement. T r e a t m e n t plan requires a multidisciplinary approach with input f r o m all involved surgical specialties and the primary care physician. We believe that these children should be treated at tertiary referral centers because of the extensive support needed for the multisystem failure and subsequent c o m p l e x treatment modalities.

CONCLUSIONS

5

small vessel thrombosis, thereby limiting tissue loss due to the ongoing purpura. P r e m a t u r e discontinuation of heparin during persistent t h r o m b o c y t o p e nia has been associated with signs of intensified circulatory failure in the lower extremities of children and further tissue loss [5]. The a b s e n c e of palpable peripheral pulses and sluggish capillary refill in the extremities is most likely a reflection o f hypotension, s e c o n d a r y to septic shock, which was present in the children we saw early in the disease process. Adequate fluid resuscitation and cardiopulmonary support is m a n d a t o r y in the early phases of the disease. Although the distal extremities can be gangrenous without detectable pulses, there is no need for arteriography, as the underlying pathological process is one of small vessel thrombosis and sepsis, not large vessel disease. Heparin should be administered in a loading dose of I00 units/kg followed by a continuous heparin drip to maintain the partial thromboplastin time at two times normal. The heparin should be continued until coagulation factors and platelet counts in the pre-heparin blood specimen have returned to normal levels. Transfusions of red cells may be required, but, specific factor replacement is not effective in reversing the DIC. There is no evidence that steroids or hyperbaric oxygen are beneficial, Dextran may be a useful adjunct but requires further trials. The operative treatment of these patients is limited to amputation of devitalized and necrotic tissue while trying to c o n s e r v e limbs and functional fingers and toes. In the five largest series reported within the last 15 years, 20 o f the 36 patients had major amputations (below-knee amputation, aboveknee amputation, a r m amputation), and all patients had some degree of tissue loss [1,2,4,7,10]. Our

mmm

In conclusion, purpura fulminans is a rare but interesting problem for the consulting pediatric, vascular and plastic surgeon. Although extremity perfusion seems poor, arteriography is unnecessary. Continuous heparin and conservative management of the purpuric lesions is the treatment of choice, until final demarcation occurs. Finally, complex plastic reconstructive procedures may be required to obtain the best possible functional results.

REFERENCES 1. DUDGEON DL, KELLOGG DR, GILCHRIST GS, WOOLEY MM. Purpura fulminans. Arch Surg 1971;103: 351-358. 2. SILBART S. OPPENHEIM W. Purpura futminans. Clin Ortho Re1 Res 1985:193:206-2t3.

3. FISHBEIN RH. Purpura fulminans: an analysis of the lesion and its treatment. J Ped Surg 1969;4:320-329. 4. CHU DZJ, BLAISDELL FWG. Purpura fulminans. Am J Surg 1982;143:356-361.

5. ANTLEY MR, McMILLAN CW. Sequential coagulation studies in purpura fulminans. N Engl J Med 1967;276: 1287-1290. 6. SPICER TE. RAU JM. Purpura fulminans. Am J Med 1976; 61:566-571. 7. CANALE ST, IKARD ST. The orthopaedic implications of purpura fulminans. J Bone Joint Surg 1984;66-A:764-769. 8. HJORT PF, RAPAPORT SI, JORGENSEN L. Purpura fulminans: report of case successfully treated with heparin and cortisone: review of 50 cases from literature. Scandinav Haematol 1964;1:169-192. 9. ALLEN DM. Heparin therapy of purpura fulminans. Peds 1966;38:211-214. 10. GOOD RA, Thomas L. Inhibition by heparin of local and generalized Schwartzman reactions. J Lab & Clin Med 1952;40:804. 11. RODRIQUEZ-ERDMANN F. Studies on pathogenesis of generalized Schwartzman reaction. I. Alterations in coagulation system during generalized Schwartzman reaction of pregnant rabbit. Thromb Diath Haernorrh 1964;12:452-461. 12. HJORT PF, RAPAPORT SI. Schwartzman reaction: pathogenetic mechanisms and clinical manifestations. Ann Rev Med 1965;/6:135-168.