SUMMARY. Resuscitation treatment is always necessary in Lyell’s syndrome because - according to the most widely accepted classification - a characteristic feature of this reaction to drugs is extensive cutaneous de-epithelialization (over 30% total body surface area). The methods for calculating the quantity of fluid to infuse are considered in this paper, as also the choice of such fluids, especially in relation to the differences from the burns disease. In particular, the absence in Lyell’s syndrome of the alteration of vascular permeability that is typical of burns and causes oedema makes it possible to administer from the beginning solutions containing colloids, such as fresh frozen plasma which both performs a reanimatory function and constitutes a specific treatment for the disease, since the immunoglobulins that it contains succeed in blocking the cell receptors of the apoptosis. Fresh plasma is administered also subsequently to the resuscitation phase, while other specific treatments (intravenous human immunoglobulin, cyclosporin A in the absence of serious hepatorenal alterations, local therapy with cultured homologous keratinocytes) block the evolution of the disease and prevent fresh evolutionary onsets and the further extension of cutaneous damage. The resuscitation formula used is presented. This is a general guideline, as all therapy has to be personalized, taking various parameters into consideration, the most important of which are the time interval between the first manifestation of de-epithelialization and commencement of infusive therapy, the patient’s age, and concomitant pathologies. As Lyell’s syndrome above all affects elderly persons already suffering from other pathologies (usually cardiovascular and renal), central venous pressure has to be monitored, although it is generally recommended to avoid using central veins because of the risk of septicaemia. The ejection fraction is a valuable indication of the risk of acute decompensation and should therefore be calculated as soon as possible by echographic assessment.
Lyell’s syndrome is a reaction to drugs that because of the presence of bullae and areas of de-epithelialization resembles a second-degree burn. There may be involvement of orificial mucosae, of organs and systems other than the skin, and signs of systemic toxicity (fever).
The most widely accepted classification (Table I) defines Lyell’s syndrome as an exfoliative form affecting more than 30% of the total body surface area (TBSA).
<% createTable "Table I ","Synthesis of differential classification criteria","; Bullous ME;SJS-ADEN 1;TEN-ADEN 2;Overlap SJS TEN-ADEN 3@;Detachment;<10%;<10%;>30%;10-30%@;Type of lesion; @;Typical targets;yes;-;-;-@;Atypical targets;In relief;flat;flat;flat@;Maculae;-;yes;yes;yes@;Distribution; @;Acral;yes;-;-;-@;Díffuse;-;yes;yes;yes @§1,5§ME = multiform erythema, SJS = Stevens-Johnson syndrome, ADEN = acute disseminated epidermal necrosis (Ruiz-Maldonado, 1985), TEN = toxic epidermal necrolysis (Lyell’s syndrome)","",4,300,true %>We thus find, even in the less severe forms, a syndrome of acute cutaneous insufficiency which, as in burns, requires early initiation of infusive therapy.
The formulae generally used for the resuscitation of burn patients are applied in the calculation of the quantity of fluid to be infused.3 It is therefore imperative to calculate the area of skin involved, which includes not only the areas that are clearly affected by the necrolytic process but also areas where instable erythema occurs.
These are zones where rubbing with a fingertip cause exfoliation of epidermis that is in an initial phase of detachment (positive Nikolsky sign).
Other skin lesions present in Lyell’s syndrome (target lesions, maculae), which may be vast enough to cover the entire body surface area, are not taken into consideration; otherwise there would be an overestimation of the damage, with a consequent water overload.
The diagram used for calculating the extent of skin damage is the same as that used in burns, assigning an age-related percentage to the various fractions of the body.3 The quantity of liquid to be infused, calculated as a function of skin involvement, has to be reduced in cases of Lyell’s by at least 30% compared to the amount that would be administered in a burn of the same extent.2 This is because of the obvious differences between the two pathologies: one probable explanation why there is less need of infusive therapy in Lyell’s syndrome is that - besides the moderate degree of vascular damage to the derma - the early onset of generalized condition of impaired vascular, which is seen in burns, is not observed in Lyell’s syndrome.
The absence in Lyell’s syndrome of severe alterations in vascular permeability eliminates the dilemma of whether or not to administer colloid-containing solutions in the early phase of the disease.
The fact such once solutions have been administered there is no loss to extravascular space means that they can be used especially in the form of frozen fresh plasma.
A dual objective is thus achieved: that of providing fluids for resuscitation (as a result of the increase in volaemia caused by plasma’s high protein content) and that of obtaining a treatment which, according to current pathogenetic hypotheses, is specific to Lyell’s syndrome (because of the elevated presence of immunoglobulins that block the cell receptors of the apoptosis).
The amount of fluid to be administered in the first 24 h is calculated using the following formula:
2/3 x percentage x TBSA x body weight (kg) x 2.5
The fluid is administered as frozen fresh plasma (4 units) alternating with solutions of Ringer’s acetate (the rest). After first 24 h the amount of fluid is adjusted in relation to the clinical response and Ringer’s solution is replaced with a 5% glucose solution, to which it is possible to add insulin and electrolytic solutions.
Frozen fresh plasma continues to be administered in the same quantities also after the resuscitation phase, above all in order to exploit its antibody content, considering that i.v. human immunoglobulins used in the therapeutic protocol are not administered for not more than 4-5 days (Table II).
Table II presents the Lyell’s syndrome treatment protocol that we use.
Table II - Lyell’s syndrome treatment protocol
The most recent general treatments (i.v IG, frozen fresh plasma, cyclosporin A) have modified the disease’s natural history by blocking its evolution and the deterioration of de-epithelialization. In the past it was necessary to re-estimate the extent of skin involvement every day, as new areas could always be affected. Local treatment of de-epithelialized areas with cultured homologous keratinocytes - which rapidly leads to re-epithelialization - has made aggressive fluid therapy a necessity in the disease’s early phases.
Two factors of some importance in the resuscitation of patients suffering from Lyell’s syndrome are:
It was only in the final decades of the last century that patients with Lyell’s syndrome began to be treated in burns centres rather than in dermatology departments - while today it is considered that burns centres are the only places where the syndrome can be adequately dealt with.
Nevertheless, the transfer of patients is not always immediate. Table III presents the findings of various researchers who underline the long delays between onset of the disease and final transfer to a burns centre. Table III also presents our own experience, which is consistent with the data reported in the literature.
<% createTable "Table III ","Time interval between onset of Lyell’s syndrome and transfer to a burns centre",";Author, year of publication,(bibliographical reference);Number of patients;Time interval (days);Range (days)@;Heinbach, 1987 (14);19;7 ± 4.7;2-17@;Engelhardt, 1997 (15);14;4.4 ± 2.7;2-12@;Sheridan, 1999 (16);10;6.6 ± 1.6;1-15@;Palermo (1979-2003);23;7.8 ± 17.6;1-90","",4,300,true %>Delay in transfer not only affects the specific, general, and local treatment of the disease but also has important consequences for rehydrating treatment. Table IV presents data regarding a recent patient of ours, a woman aged 73 yr who presented three days after the onset of a Lyell syndrome caused by allopurinol with an extent of about 30% TBSA. The patient’s serous hepatorenal condition on admission (3 December 2002) improved considerably and normalized after appropriate rehydrating therapy.
<% createTable "Table IV ","Improvement of patient’s hepatorenal condition after rehydration",";Date;Fluids infused;Diuresis (24 h);Creat. (mg/dl);BUN (mg/dl);GOT (u/l);TGP (u/l 38)@;3 Dec. 2002;150 cc/h - 3600 cc;1135 cc (17 h);3.2;75;175;184@;4 Dec. 2002;80 cc/h - 2000 cc;2020 cc;2.6;66;103;165@;5 Dec. 2002;80 cc/h - 2000 cc;1910 cc;1.5;42;24;40@;6 Dec. 2002;40 cc/h - 1000 cc;1345 cc;1.3;25;18;34","",4,300,true %>Hepatic involvement in Lyell’s syndrome is frequently observed.17-22 However, even if there have been numerous reports, it is still debatable whether a cytolytic hepatopathy (Hbs Ag negative) is an integral aspect of the disease or whether it is not in fact caused directly by the drugs, by a septicaemic process, by the condition of dehydration, by hypovolaemia, or by shock.
With regard to renal involvement, the rare but often cited cases of membranous glomerulonephritis described in the literature are not recent and, as Roujeau et al. suggest, they cannot be attributed with absolute certainty to Lyell-related alterations (also, the condition was reported to have developed some years after the acute disease).
A recent case was described by Perro in 2000, who reported anuria in a child aged 11 yr suffering from Lyell’s syndrome related to the administration of carbamazepine prescribed for a condition of epilepsy that had just been discovered. The anuria ceased after evacuation of ureteral “mud” due to de-epithelialization of the urinary tract that had caused the simultaneous obstruction of both ureters.
However, when there is renal involvement, it is usually secondary to insufficient reintegration of the intravascular volume, which is reduced owing to the losses caused by the epidermolytic process. In other words, if hypovolaemia is not corrected it can cause acute tubular necrosis with consequent kidney failure.
Table V tabulates the mean age of patients in a number of sets of case histories.
<% createTable "Table V ","Distribution by age of patient in various case histories",";Author, year of publication,(bibliographical reference);Number of patients;Mean age (yr);Range (yr)@;Heinbach, 1987 (14);19;41 ± 22;2-80@;Revuz, 1987 (24);87;45.8 ± 22.7;2-90@;Roujeau, 1990 (25);253;46.8 ± 25.5;1-93@;Engelhardt, 1997 (15);14;33 ± 34;3-99@;Murphy, 1997 (26);44;44.9 ± 3.8;0.8-90@;Palermo (1979-2003);23;61.5 ± 20.6;8-93","",4,300,true %>As can be seen, the mean age is fairly high. In the 23 patients in our study, the mean age was higher than that in all the others - 12 patients (52.1%) were aged over 65 yr. Advanced age is responsible for an increased frequency of pathologies that are often multiple and generally chronic.
Tables VI and VII present data regarding the last Lyell syndrome patient, a woman, who came to our observation. This patient, aged 81, was a manifest case of Lyell’s syndrome due to taking Diclofenac; she presented de-epithelialization in 35% TBSA and her anamnesis revealed a history of arterial hypertension, chronic renal insufficiency, and diabetes mellitus.
<% createTable "Table VI ","Fluids infused and diuresis in the first 62 h",";Time interval (h);Fluids infused;Total diuresis@;First 4 h after admittance;300 cc/h - 1200 cc;97 cc@;Next 5 h;200 cc/h - 1000 cc;285 cc@;Next 15 h;180 cc/h - 2700 cc;1040 cc@;Next 3 h;120 cc/h - 360 cc;250 cc@;Next 3 h;100 cc/h - 300 cc;350 cc@;Next 24 h;80 cc/h - 1920 cc;1355 cc@;Next 8 h;10 cc/h - 80 cc;350 cc@;Total;7560;3727 cc","",4,300,true %> <% createTable "Table VII ","Some laboratory parameters recorded in the same period","; 8 Feb. 2003;9 Feb. 2003;10 Feb. 2003;11 Feb. 2003;12 Feb. 2003@;Urea (mg/dl);114;113;213;227;229@;Creat. (mg/dl);3;2.7;2;2;2.1@;Na+ (mmol/l);140;137;144;158;166@;K+ (mmol/l);4.9;4.3;4.1;3.6;3.8@;Glyc. (mg/dl);213;276;192;301;258@;Osmol.(mmosm/l);320;316;342;377;392","",4,300,true %>The drastic reduction of infusive therapy, as seen in Fig. 6, was associated with depletive treatment (furosemide) and was motivated by the sudden onset of an episode of cardiac decompensation. A condition occurred of plasma hyperosmolarity (Table VII), due not only to hyperglycaemia and the increase in urea (in part caused by hypercatabolism) but also to the considerable increase in natraemia. From then on, the infusion was adjusted in relation to central venous pressure and, when necessary, diuresis was stimulated by the administration of furosemide. Complementary to this was the performance of an echocardiogram that indicated an ejection rate of 40%.
The elevated incidence of cardiovascular and renal pathologies in elderly patients means that sudden cardiac decompensation is one of the most frequent complications that have to be avoided in the resuscitation phase:8 venous infusion is in fact an important cause of cardiac decompensation (Table VIII).
<% createTable "Table VIII ","Some causes of cardiac decompensation with its physiopathological mechanisms",";Triggering causes;Physiopathological mechanisms@;Arterial hypertension;Increased post-load@;Tachyarrhythmia;Decreased systolic output@;Myocardial infarction;Loss of myocardial mass, arrhythmia@;Cortisone therapy;Water retention@;Anaemia;Discrepancy between flow and oxygenation@;Intravenous infusions;Volume overload","",4,300,true %>Lyell’s syndrome always requires resuscitation treatment since there is a cutaneous involvement of at least 30% TBSA.
As Lyell’s syndrome does not cause the capillary alterations typical of burns, it is possible to use from the very beginning not only Ringer’s solution but also frozen fresh plasma, which performs a resuscitatory function and is at the same time a specific treatment for the disease owing to the presence of immunoglobulins that block the cell receptors of the apoptosis.
The quantity of fluid to be infused is calculated using standard formulae for burn patients; however, such formulae are to be regarded only as general guidelines and need to be applied taking into account the time interval between the onset of de-epithelialization and commencement of infusive therapy, the patient’s age (which is usually advanced), and concomitant pathologies (especially if cardiovascular and renal).
Resuscitation therapy must therefore be as personalized as possible, with continuous monitoring, and accordingly it must be based on certain parameters (listed in Table IX) that enable us to assess the patient’s clinical condition.
Table IX - Parameters for the clinical assessment of patients receiving resuscitation treatment
One of the most important parameters is without a doubt central venous pressure, which makes it possible to perform infusion of fluids on the basis of its initial level and its variations; if necessary, it has to be measured in different parts because of the risk of septicaemia, and it is advisable to avoid central veins.
In order to personalize resuscitation, an echocardiographic assessment should be performed as soon as possible in order to calculate the ejection fraction. This is generally above 60% and of reduced base in the elderly, especially in patients with cardiac insufficiency. The ejection fraction is therefore a valuable indication of the risk of acute decompensation and suggests the use of all necessary caution in the course of intravenous infusions in elderly patients.
RESUME. Un traitement de réanimation est toujours nécessaire dans les cas de syndrome de Lyell parce que - selon la classification la plus diffuse - un aspect caractéristique de cette réaction aux médicaments est une dé-épithéthialisation cutanée étendue (plus de 30% de la surface corporelle totale). Les Auteurs considèrent les méthodes pour calculer la quantité de liquides qu’il faut infuser, comme aussi le choix de ces fluides, en particulier relativement aux différences par rapport à la maladie des brûlés. En effet, l’absence dans le syndrome de Lyell des altérations de la perméabilité vasculaire qui sont typiques dans les brûlures et causent l’œdème nous permettent d’administrer, dès le commencement du traitement, des solutions qui contiennent des colloïdes, comme le plasma frais congelé qui remplit une fonction réanimatoire et dans le même temps constitue un traitement spécifique e la maladie puisque les immunoglobulines qu’il contient réussissent à bloquer les cellules réceptrices de l’apoptose. Le plasma frais est administré aussi après la phase de la réanimation, tandis que d’autres traitements spécifiques (immunoglobuline humaine intraveineuse, cyclosporine A en absence de graves altérations hépatorénales, thérapie locale moyennant des kératinocytes cultivés homologues) bloquent l’évolution de la maladie et préviennent de nouvelles poussées évolutives et l’extension ultérieure du dommage cutané. Les Auteurs présentent leur formule de réanimation, en soulignant qu’elle est une ligne directrice générale, vu que toute thérapie doit être personnalisée sur la base de divers paramètres, dont les plus importants sont le délai entre la première manifestation de la dé-épithélialisation et le commencement de la thérapie infusive, l’âge du patient et les pathologies concomitantes. Puisque le syndrome de Lyell touche pour la plupart les personnes aînées déjà atteintes d’autres pathologies (normalement cardiovasculaires et rénales), il faut monitoriser la pression veineuse centrale, bien que, selon les recommandations de plusieurs Auteurs il soit prudent d’éviter les veines centrales à cause du risque de la septicémie. La fraction d’éjection constitue une indication très utile du risque de la décompensation aiguë, et il faut la calculer aussitôt que possible moyennant une évaluation échographique.