ACUTE QUADRIPLEGIC MYOPATHY WITH MYOSIN-DEFICIENT MUSCLE FIBRES AFTER LIVER TRANSPLANTATION

Defining the Clinical Picture and Delimiting the Risk Factors¹

Abstract

Background. 

In the last few years, rare cases of acute quadriplegic myopathy (AQM^*) with myosin-deficient muscle fibres occurring after solid organ transplantation has been reported. The aim of the present study was to review all cases of AQM with myosin deficient fibres seen at our institution among a large series of patients after orthotopic liver transplants (OLT), with special attention to clinical aspects and associated risk factors. Additionally, an extensive review of all ultrastructurally demonstrated cases of AQM in transplant recipients is also included.

Patients and Methods. 

Among patients involved in 281 consecutive liver transplant procedures performed in a 4-year period, 3 men and 1 woman developed an arreflexic, flaccid quadriplegia in the immediate postoperative period of OLT. After ruling out other causes of weakness, a muscle biopsy was performed and a loss of thick (myosin) filaments was confirmed by ultrastructural analysis in all cases. Accurate clinical, epidemiological, and evolutive data were recorded.

Results. 

Corticosteroids had been used at usual dosage given to liver transplant recipients; all four patients had several intra- and postoperative complications leading to receiving significantly higher amounts of hemoderivates, to develop renal failure in all cases, and to require a significantly higher number of reoperations within a few days after transplantation than our contemporaneous global series of liver transplant recipients. AQM patients required a significantly longer intensive care unit and hospital stay. Muscular recovery was the rule, but currently a mild myopathic gait remains in three patients. These and other reported cases of AQM do not histologically and clinically differ from AQM seen in other critically ill patients who have not had transplants.

Conclusions. 

Patients with a complicated intra- and postoperative course of OLT who develop newly acquired acute muscle weakness should be suspected as having acute AQM with myosin-deficient muscle fibres. In this setting, differential diagnosis with other causes of weakness should be carried out, because the prognosis of this myopathy is good with early muscle rehabilitation therapy.

Acute quadriplegic myopathy (AQM^*), a rare complication of treatment with high doses of corticosteroids in most cases associated with neuromuscular blocking agents (NMBA), was initially described as affecting asthmatic patients requiring mechanical ventilation ^(1-3). Although the precise onset of acute weakness is often difficult to pinpoint because of NMBA and sedative administration, patients typically present acute flaccid weakness immediately after cessation of such drugs and the weakness is clearly evident when the weaning from mechanical ventilation is attempted. Creatine kinase (CK) levels may be normal and electromyogram results (EMG) usually suggest the diagnosis of myopathy. The histological hallmark of AQM is a selective loss of thick (myosin) filaments; this deficiency is suggested when a skeletal muscle biopsy specimen is studied by histological ATPase reaction, and confirmed through ultrastructural microscopic analysis. Improvement of the myopathy over weeks to months after discontinuation of corticotherapy and resolution of status asthmaticus is the rule ⁽³⁾.

However, it has been observed that critically ill patients with diseases other than asthma are also at increased risk for developing this complication ^{(2, 4, 5)}. In this sense, rare cases of AQM appearing after solid organ transplants have been reported increasingly during the few last years ^(5-10); the incidence of this myopathy has been reported as 7% in the only prospective study carried out in liver transplant recipients ⁽⁹⁾. Nevertheless, whether this syndrome is the same as that observed in other critically ill patients is a controversial issue ⁽⁶⁾. In addition, the clinical picture and risk factors for this subset of patients have not been fully defined. Therefore, the aim of the present study was to review all cases of AQM with myosin deficient fibres seen at our institution among a large series of patients after orthotopic liver transplants (OLT), with special attention given to the clinical aspects and associated risk factors. Additionally, after an extensive review of the literature, we have compared all ultrastructurally demonstrated cases of AQM in transplant recipients with those observed in the classical form in critically ill patients who have not had transplants, to find possible differential characteristics in AQM affecting these two subsets of patients.

PATIENTS AND METHODS

Patients. From 1994 to 1997, 281 OLTs were performed at our institution, an urban, university, tertiary care hospital that provides medical attention for about 500,000 individuals. All patients were treated according to a well established protocol including the immediate preoperative evaluation of candidates and the anesthetic, surgical and intensive care aspects of liver transplant management ^(11-17). Patients developing an overt clinical picture of acute weakness in the postoperative period of OLT underwent an accurate neurological examination. On the basis of the clinical findings, complementary explorations were indicated to rule out neurological causes of weakness. Finally, skeletal muscle biopsies were performed only on those patients in whom an acute myopathy was suspected.

The clinical charts of patients in whom AQM with myosin-deficient muscle fibres confirmed by ultrastructural analysis was diagnosed were extensively reviewed, and data reflecting all the perioperative events, pharmacological and transfusional therapies, and further follow-up were recorded to analyze the characteristics of the clinical picture and to detect possible risk factors for developing AQM. The severity of the underlying liver disease leading to OLT was estimated according to the Child-Pugh score ⁽¹⁸⁾. The global status of patients on admission to the intensive care unit (ICU) was evaluated by using the "Acute Physiology And Chronic Health Evaluation" (APACHE-II) score ^{(19, 20)}. Finally, the initial functional state of the graft was determined according to our own classification ^{(21, 22)}.

All patients included in the present report have been followed up to now in the outpatient clinic and has been examined by the same member of the transplant team who specifically evaluated the functional muscular status at each visit.

Methods. Two specimens of deltoid or quadriceps muscles were obtained by an open biopsy under local anesthesia. One specimen was routinely processed for histochemical studies as reported elsewhere ^{(23, 24)}. Briefly, they consist of histological staining (hematoxylin & eosin, modified Gomori's trichrome, periodic acid-Schiff [PAS], oil red O) and histochemical reactions (nonspecific esterase; ATPase at pH 4.3, 4.6, and 9.4; acid phosphatase; reduced form of nicotinamide-adenine dinucleotide tetrazolium reductase; cytochrome c oxidase; and succinate dehydrogenase) performed on 6 μm thick sections from frozen tissue. The other specimen was fixed in 2.5% glutaraldehyde in phosphate buffer, postfixed in 1% osmium tetraoxide, dehydrated in ascending ethyl-alcohol, and embedded in epoxiresine (araldite durcoparTM). Then ultrathin sections (about 50 nm) were contrasted introducing uranil acetate and lead cytrate and they were examined with a transmission electron microscope (Jeol JEM-1200EXII).

The results observed in our four cases of AQM were compared with those obtained in our global contemporaneous series of liver transplant recipients. Statistics were made by using the Mann-Whitney U test for comparison of quantitative data.

RESULTS

Among the patients involved in the 281 OLTs performed during the study period, 4 patients developed an overt clinical picture of generalized weakness of acute onset shortly after surgery. The histopathological examination of muscle biopsy specimens was diagnostic for AQM with myosin-deficient muscle fibres in all cases, representing an accumulated incidence of 1.4% for this kind of myopathy over the study period.

The main clinical data reflecting the preoperative condition, perioperative events, and pharmacological and transfusional therapies in these four patients are summarized in Table 1. As shown, there was an important variability in both the type and the severity of the underlying disease indicating OLT. Three patients had mild hyperglycaemia at the time of OLT, but none was receiving pharmacological treatment for diabetes. Although only two patients had a massive intraoperative hemorrhage, all four required reoperation because of uncontrollable postoperative bleeding; one patient required an emergency retransplant because of a primary nonfunctioning graft. For this reason, intraoperative and intra-ICU transfusional requirements were significantly higher in these four patients than in our global series for packed red blood cells (46±22 vs. 13±11 units; P<0.01), fresh frozen plasma (12,750±7,084 vs. 4,653±4,091 ml; P<0.01), and platelets (63±48 vs. 12±10 units; P<0.01), as were the number of surgical procedures (3.0±1.1 vs. 1.2±0.4, P<0.001) and the total surgical time (900±273 vs. 453±131 min, P<0.001). These facts explain the doses of NMBA received by these patients, because NMBA were administered almost exclusively during surgery. All patients had hypovolemic shock requiring high volume expansion and vasoactive drugs (dopamine, dobutamine, and/or norepinephrine) during the first days after the transplant, and, accordingly, all developed acute renal failure. Although all these patients needed hemodialysis, only two underwent this procedure before the development of clinical signs of acute myopathy. Despite three patients being treated with wide spectrum antibiotic therapy for developing pneumonia, no patient received aminoglycosides. All patients received sedatives (propofol, midazolam, and/or morphine sulphate) while using mechanical ventilation, and they required high doses of insulin to control hyperglycaemia. Because no patient had rejection episodes, the dosage of steroids was that usually employed to prevent graft rejection, but in one patient the tapered doses were readministered because of retransplantation on the 3rd day. Only one patient received cyclosporine-A before signs of myopathy appeared. Graft initial function was poor in all but one patient. In these three patients, extensive ischemic hepatitis (massive in the explanted nonfunctioning graft) was demonstrated in the histopathological examination. However, after a variable length of time with significant cholestatic disturbances, good graft function was recovered during hospitalization. According to all the above detailed data, our AQM patients showed a significantly poorer APACHE-II score than that observed in our global series of OLT patients (26±3 vs. 19±6; P=0.02).

Table 1: Main clinical data from patients in the present series until the time of clinical suspicion of myopathy

The clinical, analytical, and evolutive aspects of the myopathy of the four patients with AQM with myosin-deficient muscle fibres are presented in Table 2. Clinical suspicion of acute myopathy appeared very soon after transplantation. However, it is important to note that it was not detected until sedative administration was stopped, to attempt weaning from mechanical ventilation, representing the first time a complete neurological examination could be performed. In fact, in all patients, the first clinical sign noted was hypoventilation secondary to low muscular performance leading to weaning failure or a need for reintubation. At this time, clinical examination of our four patients showed flaccid quadriplegia or quadriparesia with diffuse muscular weakness (involving the neck and facial muscles in three patients) and abolishment of osteotendinous reflexes. It is interesting that weakness did not show further impairment after detection in any patient. Although there was a variable delay from clinical weakness to serum CK determination, it was increased in three of four patients. In the first patient in whom AQM was diagnosed, the EMG showed a predominantly proximal myopathic pattern; the muscle biopsy specimen being diagnostic for AQM with myosin-deficient muscle fibres. After this patient, we decided to directly indicate muscle biopsy in all patients for whom there was clinical suspicion of acute myopathy. As seen in Table 3, the epidemiological, clinical, and evolutive data of transplant recipients in whom the loss of thick (myosin) filaments was ascertained ultrastructurally are very similar to those of other critically ill patients developing AQM.

Table 2: Neuromuscular findings of patients in the present series

Table 3: Comparative epidemiological, clinical, and evolutive data from transplant recipients (^5,6,8,9, and the present reported cases) with acute quadriplegic myopathy in whom loss of thick myosin filaments was ultrastructurally confirmed, and data from other critically ill (nontransplant) patients ^(2-5,41).

Histological findings were similar in all cases. An increased variability in the fibre size coexisting with angulated, atrophic fibres affecting either type I and type II, were clearly evident. There were no positive fibres not nonspecific esterase (except in one patient, in whom a few isolated fibres had reacted mildly) or fibre grouping indicating denervation or reinnervation, respectively. Sarcolemma of several fibres has a festooned appearance, suggesting a loss of sarcoplasmic volume (Fig. 1). Occasional myonecrosis was seen in one patient. In no patient was either significant oxidative or mitochondrial abnormalities, or glycogen or lipid deposits demonstrated. In all cases, ATPase reaction at pH 9.4 disclosed the existence of several nonreactive fibres, demonstrating the absence of ATPase activity in such fibres (Fig. 1).

Figure 1: Light microscopy photomicrographs corresponding to skeletal muscle from patient 3 of the present series. (Top panel) Hematoxylin & eosin staining shows fibre atrophy with a festooned sarcolemma suggesting loss of sarcoplasmic volume. (Lower panel) ATPase reaction at pH 9.4, which demonstrates the myosin activity, confirms that fibre atrophy affects both type of fibres, and discloses nonreactive muscle fibres (stars) suggesting a loss of myosin thick filaments.

In all patients we observed isolated features of necrotic changes by electron microscopic analysis, but we did not observe significant lipid deposits, glycogen storage, or rod or cytoplasmic bodies. Capillaries had a normal morphological appearance. In some fibres, we found patchy (three patients) or diffuse (one patient) areas with a clear loss of thick filaments preserving Z-disk and thin filaments (Fig. 2).

Figure 2: Electron microscopic micrographs corresponding to the skeletal muscle of patient 1 of the present series. (Top panel) Focal derangement of the sarcomeres (arrows) surrounded by sarcomeres of normal appearance. Note the disappearance of M-lines and the loss of thick (myosin) filaments (original magnification × 5,000; bar represents 2 μm). (Lower panel) Completely deranged myofilaments with nearly complete absence of thick (myosin) filaments (original magnification × 10,000; bar represents 2 μm).

All patients underwent vigorous physical therapy and showed progressive recovery of muscle function, allowing a successful weaning from mechanical ventilation, discharge from the ICU, and discharge from the hospital after 30±18, 45±21, and 75±30 days, respectively; these figures were significantly higher than those obtained in our global series of transplant recipients (0.7±0.8, 7±6, and 25±10 days, respectively; P<0.001 for all parameters). At discharge from the hospital, all patients had a mild to moderate myopathic gait. After a mean follow-up of 27±20 months, all patients have improved their muscular performance; three have retained a mild myopathic gait.

DISCUSSION

Although central nervous system complications after OLT are well recognized in the literature ^(25-28), studies of neuromuscular complications are rare. We report four well-documented cases of AQM with myosin-deficient muscle fibres during the early postoperative period of OLT, representing an accumulated incidence of 1.4% in our 4-year series. This figure comes from a nonprospective analysis, and it is very similar to that reported in the other retrospective review from the Mayo Clinic, in which only 7 of 520 (1.3%) liver transplant recipients developed a severe myopathy ⁽⁷⁾. However, the real incidence of this syndrome is higher than that noted in such retrospective analyses, because the only prospective study, performed by Campellone et al. ⁽⁹⁾, reported an accumulated incidence of 7% for AQM with myosin-deficient muscle fibres in liver transplant recipients of the University of Pittsburgh Medical Center.

Clinical features, morphological changes in skeletal muscle, and the muscular outcome of our patients with AQM were very similar to those observed in previous patients having either lung, heart, or liver transplants ^(5-10). Nearly all the patients reported received corticotherapy associated with NMBA. The distribution of the weakness was predominantly proximal or diffuse, mild raised serum CK was present in nearly half the patients, and loss of thick (myosin) filaments was either patchy or diffuse. Recovery, although usually slow, was the rule, and the mortality rate was low and related to the primary disease (Table 3). All these aspects confer a quite stereotypical clinical picture to the syndrome, which should lead to its early recognition. Al-Lozi et al. ⁽⁶⁾, who were among the first authors describing this myopathy in three lung transplant recipients, believed that AQM in transplant recipients could constitute an entity clinically and pathologically distinct from previously described AQM of other critically ill patients. They based this conclusion on the lack of a prolonged period of inactivity or neuromuscular paralysis preceding the syndrome (exposure to NMBA in their patients was, at the most, brief and by minimal dosage); the absence of cranial nerve abnormalities other than mild facial weakness; and a light microscopic examination revealing diffuse loss of myosin in affected, usually small fibres (in contrast to patchy distribution of myosin deficiency, usually within relatively normal size muscle fibres, seen in other critically ill patients). However, although some little peculiarities could be present in this subset of patients, in view of the clinical and histological findings of our patients and others (Table 3), it seems more reasonable to interpret all these muscular conditions as minimal variants of the same nosological entity rather than different diseases.

When risk factors for developing AQM are analyzed, only the administration of high doses of corticosteroids and the use of NMBAs have shown a strong relationship with the occurrence of AQM with myosin-deficient muscle fibres in critically ill patients in general ^(1-3) and in liver transplant recipients in particular ^{(7, 9)}. In this sense, Campellone et al. ⁽⁹⁾ found that corticosteroid dosage was a risk factor independent of illness severity in liver transplant recipients, but they were not able to evaluate the role of NMBAs because the number of different drugs and doses used in their patients were difficult to compare. It is thought that NMBAs may act as a predisposing factor for AQM because of their denervative effects on muscle mass ^(29-32). Accordingly, some authors recommend restriction of the use of such agents and/or selection of those with extrahepatic metabolism and elimination when given to liver transplant recipients ⁽⁹⁾. However, in our patients with AQM, NMBAs were almost exclusively used during surgery, indicating that, in surgically complicated patients, this preventive policy is more theoretical than applicable (Table 1). Regarding steroid therapy, our patients received corticosteroids at standard dosages to prevent liver rejection (Table 1). Moreover, signs of myopathy appeared the third posttransplant day in one patient when she was given the initial tapered doses of methylprednisolone. In addition, the clinical signs of acute myopathy did not worsen after the first clinical suspicion of myopathy and muscle improvement was observed in all cases, although corticosteroid therapy was not discontinued. In contrast with Campellone et al. ⁽⁹⁾, our data strongly supports the contention that the conventional dosages used in liver transplant recipients, without additional increases, are high enough to lead to the development of AQM. Evidently, this does not contradict the possibility that the greater the doses of corticosteroids, the higher the potential risk of developing AQM. In addition, it should be noted that withdrawal of corticosteroid therapy is not required to achieve muscular recovery. Therefore, corticosteroids, although necessary, do not seem to be sufficient for the development of AQM in transplant recipients in the absence of other risk factors.

Regarding risk factors for AQM other than corticosteroids doses, our patients fulfilled all the other conditions found by Campellone et al. ⁽⁹⁾; because they were critically ill as indicated by a poorer APACHE-II score on the first ICU day, all showed high glucose blood levels on the 7th day, and all developed acute renal failure (Table 1). However, it is important to point out that two patients had clinical signs of AQM before starting dialysis, suggesting that renal failure itself rather than hemodialysis acts directly and/or indirectly as a possible predisposing factor. Nonetheless, renal failure itself is not a condition sufficient to develop this syndrome, because, in people with acute or chronic renal failure, AQM has never been described in the absence of other predisposing factors. However, and in view of our results, it can be suggested that neither tacrolimus nor cyclosporine (three patients had not received either) nor graft dysfunction (myopathy is not a recognized complication associated with acute or chronic liver diseases leading to either severe parenchymal and/or cholestatic disturbances) contributes to the development of AQM ^{(33, 34)}.

Although this feature has not been previously reported in AQM after liver transplantation, the massive postoperative bleeding complications and the subsequent very high transfusional requirements of our patients deserve special attention (Table 1). To our best knowledge, the anticoagulants and preservants used in hemoderivates are not known to be myopathic. Therefore, if high transfusional requirements play a role, other mechanisms must be involved. Sudden and repeated changes in extracellular electrolyte levels, specially ionized calcium, may be citrate mediated, and pH during massive haemorrhage and blood replacement therapy may make the muscle mass more susceptible to myopathic agents. Moreover, a relative muscular ischemia resulting from either acute anaemia, hypovolemic shock, vasoactive drugs, or microcirculatory disturbances associated with the systemic inflammatory response syndrome (SIRS) present in these patients may further potentiate this hypothetical increased muscle susceptibility. In this setting, all these potential myopathic adverse factors (i.e., corticosteroids, NMBA, uremia, dyselectrolytemia, hyperglycaemia) may achieve their maximum expression. Finally, it should be kept in mind that these patients have different conditions (corticotherapy, surgical procedures, shock, renal failure, sepsis, SIRS) which are well recognized as extremely catabolic ^(35-41). Therefore, an enhanced muscle protein catabolism cannot be excluded as a possible contributing mechanism for developing AQM ^{(42, 43)}.

In summary, physicians caring for transplant recipients should suspect the presence of AQM with myosin-deficient fibres when unexpected muscle weakness becomes evident after discontinuing sedation and neuromuscular blockade or when the patient cannot be weaned from mechanical ventilation. Differential diagnosis of newly acquired acute weakness in the setting of the ICU is extensive and sometimes difficult, because AQM may result from a wide spectrum of diseases such as brain stem infarction, central pontine myelinolysis, critical illness polyneuropathy, prolonged neuromuscular blockade, or AQM, which usually imply quite different prognoses ^(44-46). Additionally, more seriously ill patients requiring multiple reoperations and massive transfusions, with shock, acute renal failure, and altered glucose metabolism are at special risk to develop AQM. Therefore, establishing an early diagnosis of AQM in the ICU and indicating a muscle biopsy as soon as myopathy is suspected are important because awareness of this syndrome will avoid erroneous diagnosis and allow a more accurate prognosis to be made and allow rehabilitation exercises to be initiated as soon as possible.