Safety of Live Liver Donation by Individuals With G6PD Deficiency

Initial Results and Comparative Study

Abstract

Background 

G6PD deficiency (G6PDd) is the commonest genetic enzyme defect in the world. However, baring a single case report, there is no published literature regarding the safety of donor hepatectomy in G6PDd individuals.

Methods 

Potential donors with World Health Organization class III or class IV G6PDd without evidence of hemolysis were evaluated for donation, if there was no other suitable donor. Postoperatively, donors were closely monitored for hemolysis and medications, which can induce hemolysis, were avoided. Outcomes of our first 14 G6PDd donors are presented. Postoperative course of these donors was also compared with a matched cohort of 30 non-G6PDd donors.

Results 

There were 9 left lateral segment, 2 left lobe, and 3 right lobe donors. Two G6PDd donors had biochemical evidence of postoperative hemolysis not needing any specific treatment. Postoperative liver function tests, intensive care unit stay, hospital stay, and morbidity (greater than Clavien II) were similar in the G6PDd and non-G6PDd donor cohorts. Donors in the G6PDd group had lower trough hemoglobin in postoperative period (P = 0.006), greater drop in postoperative hemoglobin (P = 0.007), and a higher need for postoperative blood transfusion (4/14 vs 2/30, P = 0.071).

Conclusions 

This is the first case series reporting the safety of liver resection in G6PDd individuals. Hepatectomy in G6PD-deficient donors is associated with a greater drop in postoperative hemoglobin and a marginally increased need for postoperative transfusion. Use of these donors can be considered with caution, and it should not be an absolute contraindication for live liver donation.

In Brief

The authors present the first experience using subjects with G6PD deficiency for live liver donation (LDLT). Fourteen patients successfully underwent donor hepatectomy with good graft function in the recipients. This experience is important because of the high prevalence of G6PD deficiency in the Middle East and South Asia, areas in which LDLT is increasing. This innovation needs to be considered with caution due to the fragility of these donor candidates in order to proceed with safety.

Living donor liver transplantation (LDLT) remains the mainstay of treatment for patients with end-stage liver disease in countries where there is no realistic chance of a deceased donor. Potential living donors undergo a battery of tests to assess their suitability for living donation to identify factors that can complicate the donor's surgery or postoperative course.

G6PD deficiency (G6PDd) is the most common genetic human enzyme defect in the world affecting over 400 million individuals worldwide.¹ It is an X-linked, hereditary genetic defect with variable clinical presentation. G6PD deficiency is more prevalent in the Arabian Peninsula, Mediterranean, and African regions and parts of India. Chronic hemolysis or acute hemolytic episodes triggered by fava beans, certain drugs and stress such as surgery or infection is the main presentation in these patients. However, a significant number of these individuals may be asymptomatic and do not develop significant hemolysis throughout their lives.²

There is limited literature regarding the risks of anesthesia and major surgery in the presence of G6PDd.^3,4 Most reports have discussed the importance of avoiding particular medications and have stressed on the importance of avoiding infections. A PubMed search could not find any published series of patients with G6PDd undergoing liver resection. An early report from Kings College discussing the evaluation of potential liver donors had included G6PDd as a contraindication for liver donation.⁵ Till date, there has been a single case report by Goralczyk et al⁶ who reported a successful LDLT from a G6PDd donor. The lack of literature in this area is probably due to the fact that this problem is largely localized to the Middle East, Africa, and South Asia which until recently have had limited access to LDLT. This situation is changing with an increasing number of centers developing in South Asia and the Middle East. We have encountered a number of potential donors who are G6PDd, and our early practice was to turn them down due to the perceived risks of perioperative hemolysis. In late 2013, we started a pilot project of careful utilization of G6PDd donors for LDLT in the absence of any other suitable family donor. We present our series of 14 consecutive donors with G6PDd who underwent live liver donation.

METHODS

All potential donors underwent G6PD testing as part of our donor assessment protocol. An initial qualitative testing was used to identify the presence of G6PDd. If confirmed, a quantitative testing was then carried out to assay the amount of functional enzyme (in terms of mg% and % of normal). Donors were then evaluated for evidence of chronic hemolysis. Donors with World Health Organization class I or II deficiency were not considered for donation⁷ (Table 1). Donors with G6PD class III (G6PD activity, 10% to 60% of normal) or class IV deficiency with subnormal function (activity >60% and <100% of normal) underwent further donor evaluation only if no other suitable family donor was available. A detailed discussion with the donor regarding the need for additional perioperative monitoring, risk of perioperative hemolysis, and possible need for blood transfusion was completed before completion of assessment.

TABLE 1: WHO classification of G6PD deficiency

Perioperative care for donors with G6PDd was similar to standard liver donors with additional emphasis on sensitizing the anesthetic and medical care team regarding the medications to be avoided. All patients received Inj Amoxicillin-clavulanic acid or Inj Cefoperazone-sulbactum as perioperative antibiotic prophylaxis and intravenous N-acetyl cysteine in the first 72 hours. Parenchymal transection was carried out using Cavitron Ultrasonic Suction Aspirator (Cavitron Integrated Systems, York, PA) without inflow occlusion. The same donor team carried out all operations. At the end of the procedure, a single drain was placed in all donors along the cut surface. Postoperative analgesia was achieved by epidural analgesia. Acetaminophen was avoided.² A high index of suspicion for any symptoms of infection or acute hemolysis and regular laboratory monitoring for sepsis and hemolysis was maintained. Screening for hemolysis was undertaken from days 1 to 5 postoperatively and then as required (Figure 1).

FIGURE 1: Protocol for evaluation and perioperative management of G6PD-deficient liver donors in our program.

For the purpose of this study, outcome data for all G6PDd liver donors who underwent donor hepatectomy between September 2013 (when we set up the protocol for liver donation by G6PD-deficient donors) and May 2015 is presented. To compare outcomes with donors without G6PDd, a matched cohort was prepared matching each study donor with at least 2 control donors matched for age (20-30 years, 30-40 years), sex, donor weight (<60 kg, 60-75 kg, >75 kg), and type of graft (left lateral or nonleft lateral). Intraoperative and postoperative variables, morbidity (in terms of Clavien-Dindo grading),⁸ and duration of stay were compared for the 2 groups. Biochemical confirmation of hemolysis was based on decreasing serum haptoglobin, increased urinary hemoglobin, and peripheral smear evidence of hemolysis. The study protocol was approved by the institutional ethics committee.

All statistical analyses were done on SPSS version 16. Data are presented as median with interquartile range. Nonparametric tests (Mann-Whitney U and χ²/Fisher exact test) were used to analyze differences in variables between the G6PDd and non-G6PDd cohorts. A P value less than 0.05 was considered statistically significant.

RESULTS

Between September 2013 and May 2015,172 LDLT (adult, 109; pediatric, 63) were undertaken in our unit. Among these, 14 patients received grafts from living related donors with confirmed G6PDd. Among the 14 G6PDd donors, 9 donors had class III deficiency and 5 donors with class IV G6PDd. The median age of the donors was 28.5 years. There were 9 male donors (64%). All donors donated to their close family members with end-stage liver disease with or without hepatocellular carcinoma. There were 9 left lateral donors, 2 left lobe donors, and 3 right lobe donors. The preoperative evaluation, intraoperative events and postoperative course of the donors is detailed in Table 2. Twelve of these donors had no additional comorbidities, 1 donor had class III G6PDd and sickle cell trait, whereas another donor had class IV G6PDd and protein S deficiency.

TABLE 2: Preoperative and postoperative course of 14 G6PD-Deficient donors in chronological order

Perioperative Course

There were no significant intraoperative complications in any of the patients. No patient needed intraoperative blood transfusion. There was no evidence of intraoperative hemolysis, and all donors were extubated in theater and shifted to intensive care unit (ICU). Median ICU stay in the group was 4 days, and overall postoperative hospital stay was 9 days. Postoperatively, 1 patient has a grade 3A complication (drain site bleeding needing suturing under local anesthesia and intravenous sedation). Four patients needed single-unit packed cell RBC transfusion for decreasing hemoglobin in the early postoperative period. In 3 of these cases, ongoing hemolysis was considered a possibility though none had biochemical evidence of hemolysis. All patients were fully independent and on normal diet at discharge. Detailed information regarding the clinical course of the entire study cohort is shown in Table 1. All donors were well on follow-up with stable hemoglobin.

Matched Pair Analysis

The control group and the G6PDd group showed no difference in the operating time, blood loss, need for intraoperative blood transfusion, duration of ICU or hospital stay, and complications as scored by Clavien grading. None of these donors needed intraoperative transfusion. Two donors in the non-G6PDd group needed postoperative transfusion. The trough postoperative hemoglobin in the G6PDd group was significantly lower. The fractional drop in hemoglobin from the preoperative to the trough postoperative levels was significantly higher in the G6PDd group (Table 3). The need for postoperative blood transfusion was higher in the G6PDd group though it was not statistically significant (4/14 vs 2/30, P = 0.071).

TABLE 3: Comparison of preoperative and postoperative variables of 14G6PD-Deficient donors and 30 matched Non-G6PD–Deficient donors

DISCUSSION

Living donor liver transplantation is a necessity in large parts of the world with limited access to deceased donor liver transplantation. Although zero donor mortality is the ultimate aim in LDLT, realistic figures of donor mortality range from 1 in 200 to 500 donor hepatectomies.⁹ The aim of preoperative donor assessment is to evaluate the donor's health in detail and identify any factors that can complicate the donor's surgery or postoperative course. Some conditions identified may contraindicate donation whereas other factors may trigger specific management plans to ameliorate any additional risk.

G6PDd is an X-linked, hereditary genetic defect caused by mutation in the G6PD gene. Several mutations have been identified, producing a wide variety of biochemical and clinical variants. G6PD deficiency has a characteristic worldwide distribution with high prevalence rates in Africa, the Middle East, Southern Europe, and South Asia.¹ The geographical distribution of G6PDd prevalence is closely related to the endemicity of malaria in those populations. The incidence of G6PDd is low in Europe (<0.5%), Far East (<0.5%), and North America (0.5%-3.0%). It is highest in sub-Saharan Africa and Saudi Arabia (15%-25%). The rest of the middle eastern countries and the Indian subcontinent have prevalence rates ranging from 5% to 15%.⁷ G6PD deficiency has been reported in 2.3% to 27% of all individuals in Indian tribal populations with an average of 7.7%.¹⁰ G6PD deficiency primarily afflicts men; however, in areas with high frequency of mutation, homozygosity or double mutations in both the X chromosomes have been reported, leading to clinical manifestations of G6PDd in women also.⁷

Despite the high incidence of this condition, a PubMed search does not reveal any literature regarding the safety of liver surgery or living donation in G6PDd individuals. The Amsterdam guidelines¹¹ for the living renal donor include testing for G6PDd (where indicated) in the donor evaluation process. Baker et al⁵ in an early article discussed the evaluation of potential living liver donors and stated that G6PDd should be a contraindication for donation. The only published report of a G6PDd individual undergoing donor hepatectomy was from Goralczyk et al.⁶ Here, a 25-year-old donor of Middle Eastern origin was assessed as a potential donor for his father. The donor was clinically well with normal blood counts and liver function. He underwent right lobe donation with middle hepatic vein included in the graft. Postoperatively, the donor's bilirubin increased to 8.4 mg/dL but normalized within 1 week. The authors did not provide information regarding the severity or class of G6PDd, evaluation done, or perioperative measures taken to minimize the risk of hemolysis. The peak bilirubin of 8.4 mg% is significant, though the authors claimed that such elevation is common in their donors. It was not clear from the article whether this was due to hemolysis. The paucity of literature in this area is probably because this problem is localized to countries in the Middle East, Africa, and South Asia which until recently have had limited access to LDLT. This situation is rapidly changing, with centers in South Asia and the Middle East hosting some of the largest LDLT programs. There is hence an urgent need to have some clarity with regards to the safety of G6PDd donors.

Liver donation entails major surgery with the risk of acute hemolysis due to surgical stress, infection, or exposure to certain drugs. Although the risk may be acceptable in an individual undergoing surgery for his/her own benefit, questions remain as to whether the risk is acceptable in a living donation setting. Given our present experience, we believe that G6PDd should not be an absolute contraindication for liver donation. The transplant team should carefully evaluate the severity of the G6PDd and the risk of significant perioperative hemolytic episodes. If the risk is sufficiently low, and there are no other living donor options available, these donors may proceed with donation as long as they fully understand the risks involved. In addition, the entire transplant team including the nonmedical support staff should be educated regarding the nature of this problem. The donor should be managed by a team conversant with the condition and with a care plan developed with reference to G6PDd.

There are 2 aspects of our study that need further elaboration. Nearly two thirds of our donors in the study were for pediatric recipients. This is a reflection of our significant pediatric transplant practice (one third of all LDLT in our program) and that majority of our pediatric transplants are in patients of Middle Eastern origin. These donors had a left lateral sectionectomy, which is a relatively smaller operation as compared with a right hepatectomy. We found no evidence of biochemically proven hemolysis in these patients even though the levels of G6PD enzyme were very low. The postoperative course was similar in the 3 right lobe grafts and 2 left lobe grafts in the group. Two of the right lobe donors had significant elevation of peak bilirubin (9.4 mg% and 7.9 mg%, respectively), though both normalized rapidly. We however do believe that the cutoff limit for functional G6PD activity for right lobe donors should be higher than for left lateral donation, though we do not sufficient data here to support this contention. Currently, we would not consider a donor for right hepatectomy if the G6PD activity was less than 25%. During this study period, 2 donors being assessed for right lobe donation with class III deficiency and G6PD activity less than 25% were rejected. This cutoff is arbitrary, but this once again reflects the total lack of any literature regarding the safety of liver resection in G6PDd individuals. Literature suggests that class III deficiency (10%-60% of normal activity) is associated with the risk of limited hemolysis in response to stressor drugs and infection. We have used an intermediate value of 25% as the cutoff assuming that the risk decreases as the level increases from 10% to 60%. We hope our data can serve as a starting point for further discussion and evaluation.

Second, 4 patients in the G6PDd group needed red cell transfusion in the postoperative period. On further evaluation, the trough hemoglobin at the time of transfusion was greater than 7.0 g% in all 4 instances. Three of these patients had no evidence of ongoing hemorrhage, and the transfusion was triggered by concerns regarding continued hemolysis. All 3 donors underwent donor hepatectomy within the first 3 months of the start of accepting G6PDd donors when we probably were more anxious about postoperative hemolysis. Over the study period, we have seen that the drop in hemoglobin in the postoperative period is temporary, and it is safe to observe these donors carefully until the hemoglobin stabilizes.

Strategies, such as preoperative iron supplementation to improve the preoperative hemoglobin, may be an option in these patients if they have a low normal hemoglobin as is common in many Indian women. Preoperative self-donation may not be an appropriate option as banked red cells gradually loose the limited G6PD function over the storage period, making them more prone to hemolysis. It is hence safer to use fresh red cells from non-G6PDd donors in case these donors need transfusion instead of self-banked red cells.

There are 2 alternative arguments that we have encountered during our discussions with transplant professionals regarding the use of G6PDd donors for liver donation. One argument is that these donors have an additional risk and hence should be completely excluded. The problem is that G6PDd is not uncommon and is, in parts of Africa and the Middle East, a fact of life. Excluding donors based on G6PDd alone would mean excluding a significant proportion of the donor pool. As mentioned earlier, no unit should proceed with a G6PDd donor if an alternative donor or a deceased donor option is available. Unfortunately, G6PD is a hereditary condition, and we have encountered instances where 2 or more family members are G6PDd. From our own series, these patients who underwent transplantation would not have been transplanted in a timely fashion if we had not accepted G6PDd donors. The counter argument put forward by some centers is that G6PDd testing is not necessary because donors who have never had any symptoms are unlikely to have problems during a donor hepatectomy. This we believe is a more dangerous argument, as absence of previous problems does not always mean that the donor will not have a hemolytic episode after a hepatectomy. Donors are usually young men and women, and many would not have encountered a significant stressor in their life to demonstrate the G6PDd phenotype. By quantitative estimation of the enzyme deficiency, we are able to select a group of healthy donors with a low risk of perioperative hemolysis (class III) and at the same time put in place a specific clinical algorithm to manage these patients in the postoperative period.

In conclusion, we report the first series of 14 living liver donors with G6PDd who underwent donor hepatectomy. We have found that these donors have a steeper drop in postoperative hemoglobin, which is probably related to stress-related hemolysis and may translate to an increased need for transfusion. However, in the absence of an alternative family donor and low chance of receiving a deceased donor graft, G6PDd donors may be considered for liver donation if they are otherwise fit for donation, have a low predicted risk of perioperative hemolysis, and are managed in a vigilant clinical setting. We believe the key is to routinely test potential donors from high prevalent ethnicities for G6PDd so that informed clinical decisions can be made. G6PDd should be considered a relative and not an absolute contraindication for live liver donation.