key: cord-0005847-9j4w4nor authors: Lopez Messa, J. B.; García, A. title: Acute polyneuropathy in critically ill patients date: 1990 journal: Intensive Care Med DOI: 10.1007/bf01724794 sha: f8ce91d4361a29c2439c23cf97093de3925a6fd0 doc_id: 5847 cord_uid: 9j4w4nor We studied five patients in the intensive care unit (ICU) with acute polyneuropathy. All had previously presented severe infectious processes, accompanied by diverse organ failure accompanied by the Adult Respiratory Distress Syndrome (ARDS) in all cases. Two patients died and the three survivors suffered severe motor deficiencies. Electromyographic studies revealed axonal damage which predominately affected motor neurons. Cerebrospinal fluid was normal in all the patients except one, who showed a moderate increase of lymphocytes. The common causes of polyneuropathy were excluded, but in all cases a nutritional disorder was detected, based on laboratory values of proteins, serum albumin and transferrin. We conclude that polyneuropathy is relatively frequent among critically ill patients and must be closely monitored because of diagnostic difficulties and the repercussions on the progress of these patients. In spite of uncertainties about its cause, it appears to be related to severe infectious processes, ARDS, and nutritional disorders. The description of acute polyneuropathy in intensive care unit (ICU) patients, generally associated with severe infectious processes and Acute Respiratory Failure (ARF), is a recent occurrence [1] [2] [3] [4] [5] , and this disorder is poorly understood. In an attempt to establish the characteristics and possible factors contributing to the development of this disorder, patients in our unit with acute polyneuropathy were studied using a previously established protocol. We studied all patients admitted to the ICU for a variety of reasons. At presentation none of the patients had any neurological abnormalities but all developed the same pathological pattern, consisting of a motor deficit in the extremities and an electromyographic study compatible with the diagnosis of polyneuropathy. We excluded those patients who presented with these symptoms at the time of admission or those who already suffered from polyneuropathy. The following data were recorded: age, reason for admission, Therapeutic Intervention Scoring System (T1SS) [6] and Simplified Acute Physiology Score (SAPS) [7] results upon admission, origin and duration of the infectious process (if the patient had one), the day when polyneuropathy developed after admission to the ICU, a complete neurologicai examination performed initially and during the patient's stay in the ICU, the length of stay in the ICU, and the development of diverse organ failure. The following modified criteria were used [8, 9] . Respiratory failure: arterial oxygen tension (PaO2) less than 60 mmHg and inspired oxygen concentration (FiO2) higher than 0.4 or the need for mechanical ventilation for more than 48 h. Circulatory failure: heart failure, shock or the need for vasoactive drugs for more than 24 h. Renal failure: plasma creatinine higher than 2 mg/dL in the absence of a previous renal disease. Hepatic failure: serum bilirubin level increased by 3 mg/dL over baseline or serum transaminases levels greater than twice normal. Hematological failure: thrombocytopenia with a platelet count of less than 100000 per mm 3 or prothrombin time or partial thromboplastin time 1.5 times the normal control value. Neurological failure: coma without a previous neurological injury and in absence of sedative drugs. Metabolic failure: hypoproteinemia (serum total proteins less than 5.5 g/dL) or hyperglycemia (plasma glucose over 200 mg/dL) or metabolic acidosis. We performed the following microbiological studies: blood cultures, bronchial secretion and exudate cultures, serum antibodies against Legionella, Epstein Barr virus, Herpes zoster, Cytomegalovirus, and hepatitis B were measured. When a clinical diagnosis of polyneuropathy was made, plasma samples were taken for analysis of urea, glucose, creatinine, sodium, potassium, calcium, phosphorus, magnesium, zinc, copper, manganese, chromium, vitamins Bl2 , E, and C, total proteins, albumin, transferrin, rheumatoid factor, LE cell test, and thyroid function tests (Free T4 Index and T3 Resin Uptake). Levels of total porphyrins were determined from urine samples. A lumbar puncture with cerebrospinal fluid analysis of proteins, glucose, cells, and culture was performed. All patients underwent an electrophysiological study consisting of an electromyographic assessment of muscular activity at rest and during contraction in the muscles of upper and lower limbs using concentric needle electrodes. In addition, motor and sensory nerve conduction velocities, Fwaves, distal motor latencies and distal sensory action potentials of peripheral nerve were obtained using needle electrodes near nerve tissue and electronic averaging systems [10] . Muscle biopsies were performed on two patients and a sural nerve biopsy was done on a third patient. All patients required mechanical ventilation and Positive End-Expiratory Pressure (PEEP) for ARF and had pulmonary artery catheterization using a Swan-Ganz catheter. Nutrition was given parenterally in four cases, and enterally in one, by way of a nasogastric tube. Vitamin supplements, electrolytes and trace elements were given. Nitrogen balance was maintained throughout the treatment period. Due to the reduced size of the patients group and the absence of a control group, a descriptive statistical analysis was performed using average values and standard deviation of the parameters. Of a total of 330 patients admitted to our ICU over an eleven month period (June 1987-April 1988) five patients (4 males and 1 female) with acute polyneuropathic symptoms were studied. The study excludes a patient who presented with a clinical and electromyographic pattern typical of Guillain-Barr6 syndrome before admission to ICU. Average age of the patients was 65.0+5.6 years. The cause of admission was ARF in three cases (2 for pneumonia and 1 for thoracic trauma) and purulent postoperative peritonitis in two cases. The patient with the thoracic trauma developed severe respiratory infection a few days after admission (Table 1 ). The mean value of TISS and SAPS results on admission were 51.0_+15.2 and 14.6 +2.3 respectively. The length of the infectious symptoms, as characterized by leucocytosis and persistent fever above 38~ was 16.0+_4.3 days. The results of the neurological examination performed on all the patients upon admission were normal. The clinical diagnosis of polyneuropathy was made an average of 16.4+-4.7 days after hospitalization. All patients presented with a symmetrical paresis with hypotonia, hyporeflexia and conserved tactile sensitivity in the upper and lower limbs with a predominantly distal distribution, without disturbance of the cranial nerves. The average stay in the ICU was 37.8 _+ 6.8 days. Two patients died as a consequence of their infectious process, while the three survivors showed evidence of severe motor impairment due to polyneuropathy. The appearance of organ failure is shown in Table 2 . Four patients developed multiorgan failure (defined as the presence of organ failure in 3 or more organs). In reference to the ARF suffered by the 5 patients, catheterization of the pulmonary artery with Swan-Ganz catheters showed the following values (mean+s.d.): mean pulmonary artery pressure 25.1+2.3 mmHg, mean pulmonary capillary wedge pressure 10.6+2.3 mmHg, pulmonary shunt (Qs/0t %0) 25.4%_+3.9%. Static compliance values (mean + s.d.) were 31.6_+ 3.0 ml/cmH20. The FiO2 and PEEP levels necessary to achieve an adequate PaO2 were 0.6-+0.1 and 10.4_+0.9 cmH20 respectively. All patients showed diffuse bilateral alveolar infiltrates on the chest X-ray. These hemodynamic, gas exchange, mechanical respiratory, and radiographic values are considered diagnostic of Adult Respiratory Distress Syndrome (ARDS) [11] . All patients required prolonged mechanical ventilation, for an average of 26.6+-3.3 days. In the 4 patients in whom weaning was achieved, the procedure was difficult and prolonged, taking 7.0-+0.8 days. Table 3 . The rheumatoid factor and LE cell tests were negative in all the patients. The CSF studies showed protein and glucose values tc be normal in the five patients. The cell counts were less than 5 cells/m 3 in four patients, and 20 cells/m 3 (all lymphocytes) in the remaining case. In the electromyographic examination performed o~ each patient, we observed a severe loss of motor units with signs of active denervation as shown by the presenc~ of fibrillation potentials and positive waves in the muscles which were tested. In the nerve conduction velocit3 studies, normal distal motor latencies were observed in all patients except one. Motor and sensory conduction veloc. ities were for the most part normal or slightly slowed. Th( absence of F waves, diminished amplitude of compound muscle action potential, especially in the legs and slight anomalies in the distal sensory action potential, all suggested a process with a predomination of axonal degeneration of the peripheral nerves, most affecting the motor component [12] . The nerve biopsy showed a reduction of myelinated thick fibers with no inflammatory infiltrates, myelin ovoids, and areas of segmentary demyelination and remyelination. Both muscle biopsies revealed the absence of inflammation and signs of muscular atrophy. Prior to the diagnosis of polyneuropathy and in spite of nutritional support, the patients showed a tendency to have a negative nitrogen balance accompanied by deterioration in the studied nutritional parameters (Table 4 ). The appearance of cases of acute polyneuropathy (with predominantly axonal involvement) in ICU patients with other infectious processes and ARF has previously been described [1, 2-4, 13, 14] . There is no profound understanding of the process which can occur in up to 50% of patients with severe infectious processes lasting more than 2 weeks [15] . During our study patients with polyneuropathy constituted 117.2% of all patients admitted to the ICU with severe infectious processes, and 38.4% of those patients whose stay was over 20 days. It can be difficult to make a diagnosis when dealing with critically ill patients who have an active infectious process and are undergoing mechanical ventilation. The clinical characteristics of polyneuropathy are loss of strength and hyporeflexia of the limbs and in particular, difficulty in weaning [1, 15, 16] . Sensory involvement is variable, patients do not usually exhibit evidence of autonomic dysfunction and approximately 20% of cases have facial muscle impairment [13] . The cerebrospinal fluid glucose, protein and cell counts were generally normal, although there was a slight elevation in the protein value in some cases [1, 3]. One of our patients showed a slight rise in predominately lymphocytic cells, as has been described in other cases of polyneuropathy [t]. 161 Four of our patients, as previously reported [1, [3] [4] [5] 13] , developed an encephalopathy, seen as coma, sometimes accompanied by convulsions, before the development of the polyneuropathy. These features were reversible and may directly be related to the same central or peripheral neurological damage or be due to so-called septic encephalopathy [17] , a functional disturbance produced by toxic and metabolic disorders without a characteristic organic lesion. Although in other structural studies [14] no lesions were present in several of the cases of encephalopathy, diffuse micro-abscesses, central pontine myelinolysis, diffuse micro-hemorrhages, and multiple cerebral infarctions were encountered in others [18] . The electromyographic tests showed clear signs of axonal degeneration [1, 13] . Histological studies revealed axonal degeneration without inflammatory changes near the nerve and signs of muscular atrophy, because of denervation, as well as primary muscular changes independent of the nervous system [1, 13, 14] probably related to the septic process [14, 19] . Patients may completely recover if the primary infectious process is overcome, even though clinical and electrophysiological abnormalities continue to be present for a considerable period following the acute phase [2, 13, 14] . All the patients studied had clinical, analytic, and electromyographic data similar to cases previously described by other authors, and their nutritional abnormalities and substantial catabolic increase must be pointed out. The histological finding of reduction of myelinated thick fibers with no inflammatory infiltrates, areas of demyelination and remyelination, and myelin ovoids are typical of axonal damage. The polyneuropathy shows several similarities to the Guillain-Barr6 syndrome, although differences have been established [13] based fundamentally on electromyographic data, in which a clear pattern of axonal deterioration contrasts with the demyelinating pattern of the Guillain-Barr6 syndrome. Further differences are based on cerebrospinal fluid analysis, which shows protein elevation with albumin-cytological dissociation, while in the polyneuropathy of critically ill patients, the cerebrospinal fluid is usually normal, except for slight protein elevation occasionally. Finally, histological findings which show areas of axonal degeneration without inflammatory changes establish additional differences with the Guillain-Barr6 syndrome. In spite of these differences and because some cases of Guillain-Barr6 syndrome with an axonal pattern have been described [20] , polyneuropathy in critically ill patients could be regarded as a variant of this syndrome [21] . The appearance of a series of phenomena which affect various organs which has been termed multiorganic failure has been referred to when dealing with severe infectious processes [8] . The cause of this syndrome seems to be a disorder of the oxidative metabolism at the cellular level produced by hypoxia secondary to hypoperfusion. This produces an alteration of the basic source of energy for cellular metabolism, namely ATP. An increase in the use of ATP takes place, which causes ADP and AMP to accumulate; they are then catabolized into degraded products of purine, and, because of the hypoxia, can cross cell membranes and may directly produce cellular damage [22] . This mechanism of tissue damage to various organs in patients with severe infections could be the same occurring at the peripheral nerve level. This cellular disorder has already been demonstrated, supported by an important reduction of high energy phosphorylated components in skeletal muscle [15] . The nerve cell body is the production site for nutrients destined for the peripheral nerve axon. The axon is involved in bidirectional exchanges of nutrients between the neuronal body and distal arborescences. This cellular energy metabolism and the integrity of this flow are essential to maintain the integrity of the nerve fiber [23] . On the other hand, the fact that ARDS appears in all our cases suggests a mechanism for polyneuropathy similar to that implicated in the development of this syndrome [24] . In conclusion, cases of predominately axonal peripheral polyneuropathy in patients with prolonged infection and ARF are a relatively frequent occurrence and the cause has not been determined, although it appears to be related to one of the metabolic disorders induced by the infection. Our data in this study is consistent with other reports in associating an important nutritional disorder with these patients. In patients with severe infection accompanied by organ dysfunction, in particular ARDS, careful attention must be paid to adequate nutritional support and to controlling the catabolic state accompanying these situations. Polyneuropathy in critically ill patients Prolonged respiratory insufficiency due to acute motor neuropathy: a new syndrome Polineuropatia en pacientes s6pticos. Medicina Intensiva Polyneuropathy in the critically ill The neuropathy of the critically ill Critically ill polyneuropathy Therapeutic intervention scoring system: update 1983 Simplified acute physiology score (letter) Determinants of organ malfunction of death in patients with intra-abdominal sepsis. A discriminant analysis Fracaso Multiorg~inico (FMO). Qu6? Cufindo? Por qu6? Motor nerve conduction velocities in normal subjects with particular reference to the conduction in proximal and distal segments of median and ulnar nerve Criterios diagnosticos y etioldgicos del slndrome de distr6s respiratorio agudo del adulto The relationship of conduction velocity to pathology in peripheral nerves Critically ill polyneuropathy: electrophysiological studies and differentiation from Guillain-Barr6 syndrome Critical illness polyneuropathy. A complication of sepsis and multiple organ failure The peripheral nervous system in sepsis and multiple organ failure Weaning failure due to acute neuromuscular disease Septic encephalopathy. Etiology and management Acute haemorrhagic leucoencephalitis as a complication of the generalised Shwartzman reaction Muscle proteolysis induced by a circulating peptide in patients with sepsis or trauma An acute axonal form of Guillain-Barr~ polyneuropathy Diagnostic considerations in Guillain-Barr6 syndrome Fox IH (I985) Evidence for adenosine triphosphate degradation in critically-ill patients Does a defect of energy metabolism in the nerve fibre underlie axonal degeneration in polyneuropathies Patogenia del slndrome de distrds respiratorio agudo del adulto Lopez Messa Intensive Care Unit Hospital General