Objectives: Liver transplant recipients are at high risk of developing osmotic demyelination syndrome because of electrolyte imbalance, fluid management, transfusion, and administration of immunosup-pressive agents during the perioperative period. During the first 9 years of liver transplant procedures conducted at our institution, we experienced 5 cases (1.2%) of osmotic demyelination syndrome in 402 transplant recipients. We established a hyponatremia-oriented management protocol to carefully monitor sodium concentration during and after transplant in patients with preoperative hyponatremia. Here, our aim was to investigate the incidence of this syndrome post-transplant in patients who received hyponatremia-oriented management at our center.
Materials and Methods: We retrospectively reviewed the medical records of 1437 patients who underwent liver transplant between 2005 and 2017, after hyponatremia-oriented management had been established at our center. We evaluated predisposing conditions, clinical parameters, and biological para-meters of patients who were diagnosed with osmotic demyelination syndrome posttransplant.
Results: Of 1437 patients, 4 (0.28%) were diagnosed with osmotic demyelination syndrome based on neuroimaging findings. All 4 patients were in chronic hyponatremia pretransplant (range, 110-118 mEq/L), including 3 who experienced rapid changes of sodium (over 8 mEq/L/day) within 1 month pretransplant and 1 who had rapid increase of sodium during postoperative care. The incidence of osmotic demyelination syndrome was decreased by hyponatremia-oriented management with favorable outcome.
Conclusions: To decrease the incidence of osmotic demyelination syndrome, we should focus on iden-tifying patients at higher risk and apply appropriate management of fluids and electrolytes before liver transplant.
Key words : Electrolyte imbalance, Fluid management, Morbidity, Sodium
Osmotic demyelination syndrome (ODS) after liver transplant (LT) is a rare but potentially fatal com-plication with an incidence ranging from 0.5% to 3.5%.1-4 Although central pontine myelinolysis (CPM) is a classical presentation that reflects great susceptibility of pontine white matter tracts, extra-pontine myelinolysis (EPM) is also quite common.3 The underlying mechanism of the development of ODS remains largely unknown. However, severe pretransplant hyponatremia, rapid correction of chronic hyponatremia, and intraoperative fluctuation of sodium concentration have been reported to be common triggers of ODS in LT recipients.3-6
During the first 9 years since we started LT procedures at our center (from 1996 to 2004), we experienced 5 cases (1.2%) of ODS in 402 LT recipients.7 Patients presented with several sequelae such as speech disorders and gait disturbances.
Large changes in sodium concentration (range, 12-21 mEq/L/d) occurred during surgery for the first 3 cases. After we changed the intraoperative main fluid from isotonic saline to half saline or plasma solution A in recipients with preoperative hypo-natremia (≤ 130 mEq/L), there were no cases of acute changes in sodium concentration during surgery. However, we experienced another 2 cases of ODS, with sodium increasing over 20 mEq/L over 2 days in the intensive care unit during the postoperative period. Thereafter, we established a hyponatremia-oriented management (HOM) protocol to carefully monitor sodium concentrations during and after LT in patients with preoperative hyponatremia.7 In this retrospective study, we report the incidence, nature, causes, and functional outcomes of ODS in LT recipients after HOM was implemented at our institution based on a single-center experience.
Materials and Methods
Study patients and data collections
After obtaining approval from the Institutional Review Board of our institution (SMC 2017-07-103), we retrospectively investigated all patients who received LT between January 2005 and June 2017. We identified the incidence of pretransplant severe hyponatremia (≤ 125 mEq/L) and followed the changes in sodium concentration in these patients during the perioperative period. Characteristics of patients with and without severe hyponatremia were compared.
We confirmed the ODS by the following steps. First, recipients with medical code related to ODS and those who had preoperative hyponatremia (≤ 130 mEq), large variations of sodium concentration (≥ 8 mEq/L/d) within perioperative day 7, neuroimaging tests within postoperative month 2, or received neurologist consultation within post-operative month 2 were enrolled. Neurologic symptoms suggestive of ODS in patient medical records were reviewed, including progressive lethargy, seizure, dysarthria, flaccid quadriparesis, pseudobulbar palsy, focal neurological signs, and altered mental status. Finally, ODS diagnosis was made based on neurologic symptoms correlated with results of brain magnetic resonance imaging (MRI). We investigated the clinical and biochemical parameters of recipients with ODS up to post-operative month 1. We also examined variations of sodium concentration before LT.
Since 2005, we have performed HOM during and after LT in patients with hyponatremia. In brief, during the intraoperative period, the main fluid received during surgery is half saline or Plasma Solution A (CJ Pharmaceutical, Seoul, Korea). If the sodium concentration (arterial blood-gas analysis including electrolytes were conducted at least every hour) had increased more than 5 mEq/L from the preoperative level, additional half saline 200 mL was infused. Next, a mixture of NaHCO3 60 or 80 mEq in half saline (1 L) was infused during treatment of severe metabolic acidosis (base excess < -10 mEq/L).
During the postoperative period, a warning note was posted at bedside, such as “CPM risk! Preoperative sodium concentration is 000 mEq/L. Be careful about sudden increase of sodium con-centration.” Sodium was not mixed in the main fluid.
Statistical analyses were performed using SPSS version 23.0 (SPSS Inc., Chicago, IL, USA). The prevalence of severe hyponatremia and clinical characteristics in LT patients were estimated using descriptive statistics. For simple group comparisons, categorical variables are presented as numbers and percentages. These variables were compared using the chi-square test or the Fisher exact test. Conti-nuous variables are presented as mean and standard deviation. These variables were compared using t test or the Mann-Whitney U test. Statistical signi-ficance was defined as P < .05.
Between January 2005 and June 2017, 1437 con-secutive LT procedures (1260 adult and 177 pediatric patients, 1084 living-donor and 353 deceased-donor) were performed. Table 1 presents the incidence of severe hyponatremia (≤ 125 mEq/L) and changes in sodium concentration over 7 days during the perioperative period. Overall, 131 LTs (9.1%) were performed for patients who were in severe hypo-natremic state (≤ 125 mEq/L). Sodium fluctuation (≥ 10 mEq/L/3d) occurred in 22.9% (30/131) before surgery. However, there were no cases of rapid increases in sodium concentration during post-operative days 1 to 3. A rapid increase in sodium concentration (≥ 8 mEq/L/d) occurred in 6 recipients during surgery. Two patients were unable to recover consciousness due to impaired hepatic function of transplanted liver. Two other recipients initially recovered consciousness but died in the hospital due to sepsis. One recipient experienced hand tremor and severe postoperative psychotic delirium during postoperative month 1. However, the patient recovered without further problems. The other recipient recovered without any symptoms. However, MRI was not performed for these patients; therefore, we could not confirm the diagnosis of ODS. The 131 recipients with preoperative severe hyponatremia showed advanced liver cirrhosis and delayed recovery after LT compared with those who did not have preoperative severe hyponatremia, although survival rates were not statistically different between the 2 groups (Table 2).
Osmotic demyelination syndrome
The overall incidence of ODS was 0.28% (4/1437). All cases were adults. In pediatric recipients, ODS did not occur. Table 3 and Table 4 show the clinical findings of patients with ODS. All patients except patient 1730 achieved good functional recovery and favorable progress to date. Preoperative severe hyponatremia (range, 110-118 mEq/L) was shown in 4 patients with ODS. Three of these patients (patients 515, 831, and 881) had corrected hyponatremia with 3% NaCl before surgery (Figure 1). All 3 patients showed lethargy, dysarthria, or quadriplegia after rapid increase of sodium concentration. Although they showed altered mentality, it was difficult to differentiate from hepatic encephalopathy. The appearance of ODS lesions on conventional T2-weighted MRI was often delayed. Thus, additional imaging studies were not performed until LT.
One ODS patient (patient 1730) underwent sodium fluctuation (9 mEq/L/d) on postoperative day 3 (Figure 1). The hepatic artery was suspected to occlude. Thus, NaHCO3 was infused (NaHCO3 of 180 mEq in 5% dextrose at 1 L: hydration 3 mL/kg/h for 1 h before computed tomography and 1 mL/kg/h for 6 h after computed tomography) to prevent kidney injury following the use of contrast agent. This patient gradually developed tremor in both hands, dysarthria, and right-side weakness. An MRI examination was done on posttransplant day 29, and he was diagnosed as having CPM.
In the present study, the incidence of ODS was 0.28% (4/1437 patients), which was lower than that shown in previous studies2,3,6 and our earlier experience. Severe hyponatremia and an unintentional rapid increase of sodium concentration through fluid management in the perioperative period can cause subsequent complications in the form of ODS. Increased sodium was considered more important than severity of hyponatremia in the occurrence of neurologic complications. Lee and associates reported that patients experienced significantly higher perioperative sodium variations over 24 hours, especially during surgery, which could have been related to ODS.5 In our review, there was 6 cases (0.42%) of greater change (≥ 8 mEq/L/d) in sodium concentration following HOM during surgery. This may have been the main reason why ODS showed a low incidence in our recipients.
There are many reports about the safe rate for sodium correction.5,8-12 We identified patients based on the lowest correction rate that was reported for finding ODS cases.13,14 A diagnosis of ODS was based on neurologic symptoms correlated with results of brain MRI after LT. However, one recipient with EPM findings on MRI presented with a normal sodium level without manifestation of sodium concentration fluctuations during the perioperative period. He presented with seizure and drowsy mental status after air embolism originating from a disconnection of central venous line at the intensive care unit on posttransplant day 7. Before this event, he did not show any neurologic symptoms, although his blood pressure was not well-controlled (systolic blood pressure of about 150-200 mm Hg). Other possible explanations of ODS include the use of tacrolimus15-17 and low cholesterol concentration (26 mg/dL on posttransplant day 1).5,15 However, 1 day after the embolic event, he showed recovered mental status with mild dysarthria and hand tremor. Thus, treatment with tacrolimus was continued. After 7 days, these sequelae had fully recovered. Unlike acute vascular infarction of the brain stem, clinical symptoms of ODS generally begin 2 to 7 days after sodium concentration is increased. They symptoms are progressive.18 Thus, we speculate that the air embolic event and the uncontrolled high blood pressure might have caused the EPM-like lesion shown on MRI.
Treatment of hyponatremia in patients with cirrhosis has been emphasized because the existence of hyponatremia is a major risk factor for the development of overt hepatic encephalopathy.19,20 In addition, cirrhotic patients frequently experience hemorrhage and general weakness. They frequently receive transfusion and fluids with electrolytes. Thus, sodium concentration should be carefully monitored during administration of blood products and/or hypertonic saline during preoperative care of LT recipients. In addition, ODS is frequently suspected after all other possibilities are excluded. Conventional MRI findings lag clinical manifestations of CPM.21 However, diffusion MRI can be used for identification in the first 24 hours after onset of clinical symptoms lasting up to 21 days.18,22 Thus, early diagnosis of patients with suspected CPM is now possible with strict monitoring of electrolyte concentration and diffusion MRI.
Hyponatremia is a frequent complication in patients with advanced cirrhosis and ascites.19,23 Our review also showed that advanced cirrhosis, ascites, and hepatic encephalopathy were associated with hyponatremia in LT recipients. However, during long-term follow-up after LT, the survival rate was not significantly different between recipients with hyponatremia and those who did not have hyponatremia. Although there have been several case reports about treatment of ODS, definitive treatment that can convincingly halt the progress or reverse ODS has not been found.24 Thus, there is a need to recognize preventable adverse effects caused by the rapid increase of sodium.
Despite HOM, rapid sodium increase (≥ 8 mEq/L/d) occurred in 6 recipients during LT. All cases experienced massive bleeding during surgery and were transfused with over 10 units of packed red blood cells. In cases of massive bleeding, it is considered necessary to perform more frequent evaluations of electrolytes and modify HOM. All recipients in this review period received tacrolimus as immunosuppressant, and the 4 patients with ODS had well controlled blood pressure. However, future protocols must include the comprehensive con-sideration of these risk factors.
Our study has several limitations. Diagnosing ODS based on clinical manifestations can be difficult, as signs and symptoms of ODS are often absent or masked by other neurologic abnormalities unrelated to ODS. Findings from autopsy series suggest that asymptomatic or mildly symptomatic ODS may be significantly more common than what is suspected in LT patients.12,25 All patients with clinically possible or asymptomatic ODS did not undergo routine brain imaging. Thus, the true incidence of ODS could be higher. Because of the low incidence of ODS, we could not analyze risk factors or establish a safe rate of sodium concentration increase. However, we found that the incidence of ODS was low in hyponatremia patients who did not receive a rapid increase of sodium concentration.
We found that ODS was gradually decreased after implementation of HOM in our institution. Our review indicates that the occurrence of ODS not related to hyponatremia seems to be rare. We should focus on identifying patients at higher risk of ODS and observe them comprehensively with consistent fluid and electrolyte management by a multidis-ciplinary team to achieve favorable recovery.
DOI : 10.6002/ect.2018.0216
From the 1Department of Anesthesiology and Pain Medicine, Sungkyunkwan
University School of Medicine, Seoul, Korea; and the 2Department of
Anesthesiology and Pain Medicine, Kangwon National University School of
Medicine, Chuncheon, Korea
Acknowledgements: The authors have no sources of funding for this study and have no conflicts of interest to declare.
Corresponding author: Gaab Soo Kim, Department of Anesthesiology and Pain Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Ilwon-ro, Gangnam-gu, Seoul 06351, Korea
Phone: +82 0234100360
Table 1. Incidence of Severe Hyponatremia and Changes in Sodium Concentration
Table 2. Characteristics and Outcomes of Recipients With and Without Preoperative Severe Hyponatremia
Table 3. Details of 4 Recipients Who Developed Osmotic Demyelination Syndrome After Liver Transplant
Table 4. Intraoperative Details of 4 Recipients Who Developed Osmotic Demyelination Syndrome After Liver Transplant
Figure 1. Changes in Sodium Concentration in 4 Patients With Osmotic Demyelination Syndrome