Ann. Medit. Burns Club - vol. VIII - n. I - March 1995

A CASE OF TOXIC EPIDERMAL NECROLYSES ASSOCIATED WITH MYCOSIS FUNGOIDES AND COMPLICATED BY CONSUMPTION COAGULOPATHY

Divisione di Chirurgia Plastica e Terapia delle Ustioni, Ospedale Civico e Benfratelli, USIL 58, Palermo, Italia

SUMMARY. A case of drug-related toxic epidermal necrolysis (TEN) is described in an adult patient suffering from mycosis fungoides. The course of the disease, which affected 50% of the body surface and ended fatally following cerebral haemorrhage, was characterized by the onset of a serious hepatopathy, thrornbocytopenia and consumption coagulopathy. These alterations were considered to be TEN-specific and not secondary to the sepsis subsequently observed. Stress is laid on the complexity of this disease, which is not only cutaneous or cutaneoushnucoseal, and on the importance of immediate hospitalization in a specialized centre in order to guarantee continuity of therapeutic treatment.

The characteristic skin involvement makes toxic epidermal necrolysis (TEN) resemble a second-degree bum, with which it has several physiopathological aspects in common (e.g. loss of liquids, hypereatabolism and increased infective risk).
TEN is a reaction to drugs which manifests itself not only at the level of the skin but also in the mucous membranes of various orifices and in a number of organs and systems, sometimes primarily.
Because of this multisystemic involvement, frequently with hepatic, haematological and pulmonary alterations, TEN has a worse prognosis than a burn of the same extent. We describe here a patient suffering from TEN whom we recently observed. The patient died when skin reepithelialization was to a large extent complete.

S. G., aged 50 years, male

On 20 April 1992, because of intense bilateral gonalgia, and following a medical prescription, the patient had intramuscular injections of three phials of a piroxicambased product. Three days later he noticed the appearance of widespread, non-pruritic, confluent erythernatous patches. On 24 April 1992 he was admitted to the Palermo University Dermatological Clinic.
The patient was given systemic and topical cortisone treatment for six days and then transferred to the Intensive Care Unit (ICU) on 30 April following the onset of severe hydroelectrolytic and metabolic imbalance, after the appearance of the typical TEN skin picture.
In the ICU the patient's hydroelectrolytic equilibrium was restored and he was subjected to hyperbaric oxygenotherapy sessions, antibiotic and cortisone treatment, and total parenteral nutrition (TPN) by a central venous route (right vena subelavia).
When the patient reached our Department four days later (4 May), eleven days after the onset of TEN, he was still receiving antibiotic treatment (tetracycline) and cortisone therapy, as well as TPN.

"Patches of erythema and epidermal exfoliation in the face, neck, anterior and posterior thorax, legs and feet covering about 50% of total body surface area; the natural orifices and mucous membranes inspected are not affected" (Figs. 1, 2)

Fig. 1 - Denuded bleeding dermis in the posterior region of the thorax (pressure point). Fig. 2 - Epidermal necrolysis in both feet.

On 17 February 1992 the patient had been admitted to the University Dermatological Clinic for a skin condition from which he had been suffering for about five years. This condition had recently become intensely pruritic. The patient was given cortisonic, systemic and local treatment and discharged a few days later with a diagnosis of mycosis fungoides. The onset of TEN occurred about two months later.

The clinical parameters monitored were haematocrit, hourly diuresis, body temperature, arterial pressure and heart rate.
Haernatocrit and hourly diuresis remained normal; body temperature and heart rate were constantly high; arterial pressure was also constantly above normal, with some critical episodes requiring immediate pharmacological treatment.

Table I shows the results of the blood chemistry tests. There was constant hyperglycaernia, together, with in-creased serum enzymes and bilirubinaemia with values manifesting themselves clinically in the patient's jaundiced appearance. Plasma creatinaernia values remained normal in spite of the contemporary increase in urea.

Table II presents the results of the haematological tests. Note the constant increase in leucocytes, especially in the neutrophil portion, and the progressive and serious reduction in the number of platelets, with normal red blood cell and haernoglobin values except on the last date. Lymphocytopenia is present.

In Table III we report the patient's respiratory condition, which was monitorized by daily blood gas tests. These showed slight hypoxaemia, with one critical episode resolved by immediate oxygen therapy. Repeated thorax radiography did not show any important alterations.

Table IV presents the results of the haemocoagulation tests. Analysis of the Table enables us to evidence a normal initial haernocoagulative picture, with a hyperfibrinogenaemia compatible with the clinical condition and without signs of consumption coagulopathy or of hyperfibrinolysis. The results of tests performed two days later were however compatible with a condition of consumption coagulopathy with a reduction in the values of prothrombin activity, fibrinogen, plasminogen, antithrombin III (At 111) and an increase in APTT and FDP.

The patient initiated heparin treatment with At III concentrates. Later coagulation tests showed a stabilized picture with lengthened APTT still present, increased FDP and reduced plasminogen.

It was decided to continue antibiotic therapy
(Imipenem instead of the Rolitetracycline previously used) owing to the positivity of swabs taken from the lesions which in various areas were beginning to show signs of reepithelialization (Fig. 3) and to the contemporary administration of steroids. Because of the doubts generated by the previous disease, we began to reduce the daily dosage of steroids a few days after the patient had been admitted, with a view to their eventual elimination. Steroids had been administered for a prolonged period and it was not advisable to interrupt their use suddenly.
TPN was replaced by peripheral nutritional support as the patient was able to feed himself. This also had the advantage of eliminating the central venous access.
The patient's therapy was completed with ranitidir, vitamins, heparin and At 111, oxygen, aerosol therapy, bathing and topical treatment.
Haemocultures were performed frequently to detect any sepsis but were all negative, even if the swabs were positive. This continued until 13 May 1992 when three haemocultures were positive to Staphylococcus aureus. This positivity was reflected in the considerable increase in leucocytes (Table II).
Also on 13 May 1992 the patient had a nervous crisis, which according to the consultant psychiatrist had two possible causes:

1. a sense of persecution: the patient having once been unjustly imprisoned, in the closed atmosphere of the burns centre he was reliving his previous experience and his loss of freedom;
2. steroid psychosis.

The following day, a probable hypertension crisis caused a cerebral haemorrhage. This haemorrhage, confirmed by a CT scan, was certainly massive, also considering the altered haemocoagulative picture. The patient was transferred to Neuroreanimation and submitted to surgery, but died after a few days.

Piroxicam is a TEN-inducing drug. Out of 62 cases provoked by non-steroid anti-inflammatory drugs, Roujeau et al. (1) found piroxicam to be responsible on 13 occasions (20.9%).

As far as we know, mycosis fungoides has not been observed in association with cases related to TEN, although there have been various reports concerning lymphoproliferative diseases. An exhaustive bibliography is given by Bianchi et al. (2), who described a case of TEN in a patient with an anamnesis that said nothing of drug administration but mentioned a retroperitoneal and pelvic non-Hodgkin lymphoma discovered on autopsy.

Fever and tachycardia are part of the general picture of TEN; this is not true of arterial hypertension and hyperglycaernia, which in all probability are due to prolonged steroid treatment.

The high plasma urea levels, with normal diuresis and normal creatinaemia values, are not compatible with renal dysfunction and may be related to the condition of hypercatabolism, the acid-base equilibrium disturbances and the steroid and antibiotic (tetracycline) treatment administered to the patient.
Revuz et al. (3) believe that the level of plasma urea is an important prognostic factor because this depends on both renal function and the gravity of stress and catabolism. Other prognostic factors which, on the basis of the multifactorial analysis that we carried out, appear to be important are age and the percentage of body surface area involved.
Renal involvement, when it occurs, is generally secondary to inadequate reintegration of intravascular volume, which is reduced as a result of the losses caused by the epidermolytic process. In other words, untreated hypovolaemia causes acute tubular necrosis with consequent kidney failure.
The rare cases of membranous glomerulonephritis described in the literature (4) and subsequently cited (5, 6, 7, 8, 9) are not very recent and according to Roujeau et al. (1) cannot be attributed with absolute certainty to specific TEN-related alterations.

Alterations of transaminase and bilirubin_ are frequently reported.
In the cases observed by Kvasnicka et al. (10) three patients out of eight presented a considerable increase in transarninase. Tagami et al. (11) described a case of ampicillin-induced TEN in which the alteration of hepatic function (increased enzymes and bilirubin) continued for six months after the patient recovered. The pathogenesis of the hepatic damage.and, in this particular case, also the pancreatic damage because of the contemporary increase in amylasaemia, was in the authors' opinion identical to that of the skin damage and was secondary to the liberation by the patient's lymphocytes, when they encountered the antigen, of biologically active lymphokines.
All ten patients in the series described by Westly and Wechsler (12) presented a high level of hepatic enzymes, while five patients out of six - in one of whom the condition already existed - in the series described by Aub6ck et al. (13) presented altered hepatic function. In these last cases TEN was induced by allopurinol.
Halebian et al. (14) compared two groups of 15 patients each. One group was treated with steroids and the other was not. The alteration of the hepatic enzymes was the same in both groups (nine patients in the treated group and ten in the untreated group). Jaundice (bilirubin > 5) was clinically evident in one patient in each group. It was hypothesized that although the mechanism of the alteration was not clear it may have been due to a primary disturbance rather than to a secondary process.
Saiag et al. (15) found that hepatic enzyme levels were more than doubled in eight out of fourteen HIV-infected patients suffering from TEN.
Despite numerous reports it is still controversial whether cytolytic hepatopathy (HbsAg negative) is an integral part of the disease or whether it is not instead directly caused by a drug, a shock condition or a septicaemic process (9).

The haematological and haemocoagulative alterations described in single cases and some extensive series also require further investigtion.
A case of leucopenia associated with thrombocytopenia is reported by Anhalt and Snelling (16) and also by Fenton and English (17). A case of leucopenia associated with anaemia and thrombocytopenia is reported by Goens et al. (18) and also by Castelli et al. (19).
Westly and Wechsler (12) observed leucopenia in all ten of their patients. This alteration occurred on day 2-5 of the disease. Values returned to normal (day 8) in only three patients, two children and an adolescent not treated with steroids, who recovered from the disease. Leucopenia was considered to be a cardinal aspect of TEN and its persistence was regarded as a negative prognostic factor. Age and steroid treatment were also considered to be related to the outcome.
Lymphopenia, due to a reduction in the number of T lymphocytes, in particular T-helper lymphocytes, possibly caused by preferential consumption in an immunological conflict, was described by Roujeau et al. (20) in six cases of TEN.
Bombal et al. (21) carried out a retrospective study of 26 patients suffering from TEN and found that normochromic and normocytic anaemia and lymphopenia, which may be so marked that there is a total absence of lymphocytes detectable in the blood, are almost constant haematological anomalies.
Saiag et al. (15), in their fourteen AIDS patients, found thrombocytopenia (two patients), neutropenia (two patients) and Jymphopenia (ten patients). Except for lymphopenia these anomalies cleared up spontaneously.
In short, both the platelets and the red corpuscles or the lymphocytes or the neutrophil granulocytes, either in or not in association among them, may be altered, mainly in the sense that they are reduced.
According to Roujeau et al. (9), "anaemia is present in nearly every patient because of a number of factors, including erythroblastopenia ... lymphopenia is present in 90% of patients caused by a selective and transitory depletion of the CD4+ T-helper lymphocytes ... thrombopenia in 15%."
Revuz et al. (3) contest the theory that neutropenia has a prognostic significance independently of other factors.
Halebian et al. (14) found leucopenia with neutropenia in 13 cases. Septicaemia was also present in 12 cases. It was however pointed out that while on the one hand neutropenia is important in determining the onset of septic complications it is also true that steroid treatment may affect the leucocyte count.
Kvasnicka et al. (10) found a normal leucocyte count in two of their eight patients and leucocytosis in the other six. These researchers however had concentrated their attention on the patients' haemocoagulative condition, observing signs of disseminated intravascular coagulation (DIC) of greater or lesser extent in relation to the severity of the clinical conditions and to the moment when the tests were performed.

Respiratory system

Apart from pneumonia, which is one of the most frequent complications of TEN, the main alterations of the respiratory system consist of subclinical interstitial oedema detectable by thorax radiography and mild hypoxia shown by blood gas test (9).
For this reason Pruitt (6) recommends daily chest Xray and blood gas analysis in order to monitor the respiratory system so that intubation can be performed immediately if the necrolytic process begins to affect the mucosa of the iracheobronchial tract.

We should like to stress two main points:

1. Toxic epidermal necrolysis is a complex disease: the alterations in the case we have described (hepatopathy, thrombocytopenia, consumption coagulopathy) may be independent factors in the disease or they may be related (hepatopathy and coagulopathy); they may on the other hand be secondary to sepsis which is frequently a complicating factor in TEN. In our case we found sepsis due to Staphylococcus aureus, but only after the above alterations had occurred.
2. TEN is not usually a disease that receives continuous treatment. Halebian et al. (14) calculated the time interval between the onset.of the disease and admission to their burns centre of the two groups of patients they compared with regard to the use or non-use of steroids. Patients in the first group were admitted after a mean period of 7.3 ± 0.7 days, and in the second group after 8.3 ± 1.1 days. As admission to a specialized centre is never immediate, we find ourselves faced with methods (e.g. the use of central venous access) or therapeutic approaches (antibiotics, steroids) which may not be advisable. In our patient the steroid treatment was probably responsible for a condition of psychosis. This would not be the first case, as Halebian et al. (22) reported that one of the fifteen patients treated with steroids appeared to suffer from psychosis.

It is also probable that the prolonged steroid treatment triggered the mechanism which in a situation of coagulopathy not only favoured the onset of sepsis but also caused the patient to pass from a condition of hypertension to the acute vascular accident and finally to his death.

RESUME. Les auteurs décrivent un cas de nécrolyse épidermique toxique (sigle en anglais, TEN) dans un patient adulte atteint de mycosis fongoïde. Le cours de la maladie, qui touchait 50% de la surface corporelle et s'est terminé par la mort du patient après un épisode d'hémorragie cérébrale, était caractérisé par la manifestation d'une grave hépatopathie, thrombopénie et coagulopathie de consommation. Les auteurs, qui considèrent ces altérations spécifiques pour la TEN et non secondaires à la condition septique successivement observée, soulignent la complexité de cette maladie, qui n'est pas seulement cutanée ou cutanée/muqueuse, et l'importance de l'hospitalisation immédiate chez un contre spécialisé pour garantir la continuité du traitement thérapeutique.

BIBLIOGRAPHY

1. Roujeau J.C. ', Guillaume J.C., Fabre J.P., Penso D., Fldchet M.L., Girre J.P.: Toxic epidermal necrolysis (Lyell syndrome). Incidence and drug etiology in France, 1981-1985. Arch Dermatol., 126: 37 42, 1990.
2. Bianchi L., Gatti S., Carrozzo A.M., Marinaro C., Iraci S., Luzi F., Nini G.: Sindrome di Lyell e linforna. G. Ital. Dermatol. Venereol., 125: 255-8, 1990.
3. Revuz J., Penso D., Roujeau J.C., Guillaume J.C., Rowland Payne C., Wechsler J., Touraine R.: Toxic epidermal necrolysis. Clinical findings and prognosis factors in 87 patients. Arch. Dermatol., 123:1160-5, 1987.
4. Krumlovsky F.A., Del Greco F., Herdson P.B., Lazar P.: Renal disease associated with toxic epidermal necrolysis (Lyell's disease).Am. J, Med., 57: 817-25,1974.
5. Management of toxic epidermal necrolysis. Lancet (Editorial), 2:1250-2, 1984.
6. Pruitt B.A.: Burn treatment for the unburned. JAMA (Editorial),257: 2207-8, 1987.
7. Ward D.J., Kazeminska E.C., Tanner N.S.B.: Treatment of toxic epidermal necrolysis and a review of six cases. Bums, 16: 97-104,1990.
8. Pousa F., Valero J., Vazquez-Baffo A., Trineado S.: Bum unit treatment of three Stevens-Johnson syndrome cases with cryopreserved allograft. Ann. Medit. Burns Club, 5: 160-3, 1992.
9. Roujeau J.C., Chosidow 0., Saiag P., Guillaume J.C.: Toxic epidermal necrolysis (Lyell syndrome). J. Amer. Acad. Dermatol., 23:1039-58, 1990.
10. Kvasnicka J., Rezac J., Svejda J., Duchkova H., Kaze F., Zalud P.,Richter J.: Disseminated intravascular coagulation associated with toxic epidermal necrolysis (Lyell's syndrome). Br. J. Dermatol.,100: 551-8, 1979.
11. Tagami H., Tatsuta K., Iwatski K., Yamada M.: Delayed hypersen sitivity in ampicillin-induced toxic epidermal necrolysis. Arch.Dermatol., 119: 910-3, 1983.
12. Westly E.D., Wechsler H.L.: Toxic epidermal necrolysis. Granulocytic leukopenia as a prognostic indicator. Arch. Dermatol., 120: 721-6, 1984.
13. Aubbck J., Fritsch P.: Asymptomatic hyperuricaernia and alloponnot induced toxic epidermal necrolysis. Br. Med. J., 290: 1969-79, 1985.
14. Halebian P.H., Corder V., Madden M.R., Finklestein J.L., Shires G.T.: Improved burn centre survival of patients with toxic epidermal necrolysis managed without corticosteroids. Ann. Surg,, 204: 50312,1986.
15. Saiag P., Caumes E., Chosidow 0., Revuz J., Roujeau J.C.: Druginduced toxic epidermal necrolysis in patients infected with the human immunodeficiency virus. J. Am. Acad. Dermatol., 26: 56774, 1992.
16. Anhalt G., Snelling C.F.T.: Toxic epidermal necrolysis. Case Report. Plast Reconstr. Surg., 61: 905-10, 1978.
17. Fenton D.A., English J.S.: Toxic epidermal necrolysis, leucopema and thrombocytopenic purpura - a further complication of benoxaprofen therapy. Clin. Exper. Dermatol., 7: 277-80, 1982.
18. Goens J., Song M., Fondu P., Blum D,, Achten G.: Haematological disturbances and immune mechanisms in toxic epidermal necolysis. Br. J. Dermatol., 114: 255-9, 1986.
19. Castelli R., Centemeri D., Frattini F., Greco P.: Quadro mortale in un caso di sindrome di Lyell. Min. Anaest., 52: 345-50, 1986.
20. Roujeau J.C., Moritz S., Guillaume J.C., Revuz J., Touraine R.: Etude de sous-populations lymphocytaires an cours du syndrome de Lyell. Ann. Dermatol. Venereol., 110: 185, 1983.
21. Bombal C., Roujeau J.C. Kuentz M., Revuz J., Touraine R.: Anomalies hématologiques au cours du syndrome de Lyell. Atin. Dermatol. Venereol., 110: 113-9, 1983.
22. Halebian P.H., Corder V., Herndon D.: A burn center experience with toxic epidermal necrolysis. J. Burn Care Rehabil., 4: 176-83, 1983.