The Role Of CT‑scan Assessment Of Muscle Mass in Predicting Postoperative Surgical Complications After Renal Transplantation
Mar 16, 2022
Introduction
Renal transplantation is the reference treatment for end-stage renal failure. Compared to prolonged dialysis, transplanted patients have an average life expectancy of 9 years higher [1]. Surgical and medical complications have been assessed in the literature to evaluate for potential risk factors and consequently to optimize patient selection before renal transplantation. Thus, intrinsic morphometrics factors, such as obesity or adipose tissue distribution, seem associated with an increased risk of postoperative complications [2]. However, changes in morphometrics and homeostasis due to long-term hemodialysis need to be assessed to further predict post-transplantation outcomes. Protein-energy undernutrition is one of the known risk factors contributing significantly to dialysis mortality [3]. About 75% of hemodialysis patients on the waiting list suffer from protein-energy undernutrition [4, 5]. In 2019, the French National Authority for Health reviewed its definition of undernutrition in adults. Three phenotypical criteria are now used: weight loss, body mass index (BMI), and reduction of muscle mass or function [6]. This last criterion, which appeared in these new recommendations, introduces the notion of sarcopenia in the definition of undernutrition.
CISTANCHE WILL IMPROVE KIDNEY/RENAL FUNCTION
Although the prevalence of sarcopenia is estimated at between 5 and 37% for patients with chronic kidney disease, there are no clear data among the population of hemodialysis patients [7]. Sarcopenia is defined by a gradual and generalized muscle impairment. It is a risk factor for falls, disability or fractures, and increases the risk of mortality [8, 9]. To assess sarcopenia, there are now defnite clinical criteria involving muscle mass and function decrease (low muscle strength, low muscle quantity, and low physical performance) [10]. Computed Tomography (CT) measurement of the muscle surface area of the psoas–iliac muscles is a good tool for assessing sarcopenia because it correlates with muscle mass [11–13] and is becoming the reference measurement for nutritional and prognostic evaluation of preoperative patients in the field of oncology [14].
Yet, the associations between sarcopenia and renal transplantation postoperative outcomes have been poorly studied in the current literature especially using imaging exams. Pinar et al. have demonstrated an association between sarcopenia, calculated using psoas surface on CT scan, and 1-year post-transplantation surgical complications [15]. In their study, the authors included only obese or overweight recipients. Additionally, another study elaborated a morphometric age based on CT-scan measures and showed an association with post-operative overall survival [16]. In this study, we aimed at assessing the correlation between sarcopenia defined by muscle mass measured on CT scan and renal transplantation outcomes in a cohort of unselected recipients.
Keywords (MeSH): renal transplantation; Sarcopenia; End-stage kidney failure; Complications; kidney disease; renal failure
CISTANCHE WILL IMPROVE KIDNEY/RENAL FAILURE
Materials and methods
Study design Each consecutive patient undergoing renal transplantation in our academic hospital was retrospectively included in this study between 2011 and 2018. We excluded patients who received more than one kidney transplant, multiorgan transplantation, patients with pre-operative CT scan performed>12 months prior to the transplant or CT scan with artifact distorting the radiological measurements (intraabdominal free fluid, wall hematoma, patient movements during the CT scan). Data regarding the peri-operative period and 1-year follow-up were collected using medical notes. The length of hospitalization was collected, as well as the delay in resuming function, defined in case of need for dialysis within 7 days postoperatively. Protocols for immunosuppression, induction and postoperative maintenance were specified. The need for amines or intraoperative transfusion was also indicated.
For each included patient, muscle mass was evaluated on axial CT-scan section at the level of the third lumbar vertebra (L3) before intravenous administration of contrast (machine settings were set at 120 Kvp and used automated current modulation with a reference mAs of 400). The skeletal muscle surface (including the psoas muscles, paraspinal muscles, external oblique, internal oblique, transverse, and rectus abdominis ) was calculated using semi-automatic software. The measurements were performed by a single expert operator, using Slice-O-Matic software (version 5.0; TomoVision, Montreal, Quebec, Canada). Muscle was automatically detected between -30 and 150 Hounsfield units on the CT-scan section. This assessment calculates the skeletal muscle index (SMI) (cm2 /m2 ) by dividing the cross-sectional area (cm2 ) of the skeletal muscle at the level of the third lumbar vertebra by the square of the patient’s height (m2 ) on CT. Muscle density, in Hounsfield units, was also calculated on the same L3 CT section and represented the average density of the detected muscles (Fig. 1). As sarcopenia is defined clinically, there are no clear consensual definitions on CT scans and we could not use any validated sarcopenia definition.
The study was approved by the local ethics committee (October 2019) and was conducted following the principles of the Helsinki declaration. The database was declared to the National Board for Informatics and Freedom (Commission Nationale Informatique et Liberté, CNIL).
Post‑operative management Patients were monitored in the nephrology intensive care unit for the first 3 days postoperatively in the absence of complications before transfer to the nephrology department. Urinary catheter and drain were, respectively, removed at days 5 and 6. The JJ catheter was systematically removed 1 month after surgery in a dedicated consultation. A protocol biopsy of the transplant was performed at 3 months.
Outcomes Primary endpoint was the occurrence of postoperative major complications at 1 month and 1 year after the transplantation. Complications were classified according to the Clavien–Dindo classification and a grade 3 or higher complication according to this classification was considered as major.
CISTANCHE WILL IMPROVE KIDNEY/RENAL PAIN
Statistical analysis The statistical analyses were performed with Stata software (version 15; StataCorp, College Station, Texas, USA), considering a bilateral first species error risk of 5%. Statistical significance was set for a p value<0.05. Categorical variables were described as numbers and percentages, whereas quantitative variables as mean (±standard deviation) or median [interquartile range] with respect to their statistical distribution (normality studied by the Shapiro–Wilk test). Comparisons between independent groups for quantitative parameters were performed by Student's t-test or by Mann–Whitney test if t-test conditions were not met (normality, homoscedasticity studied by Fisher–Snedecor's test). Inter-group comparisons of qualitative parameters were performed by the Chi 2 test or by Fisher's exact test. Finally, in a multivariate situation, logistic regression was implemented by considering the covariates with respect to the univariate analysis results (p≤0.1), to study the factors associated with major complications. The results are expressed in terms of odds ratio (OR) and 95% confidence interval.
Results
Population Overall, 397 patients underwent renal transplantation during the study period whom 102 had a pre-operative CT scan less than 12 months old, and were included in the study. Of the 102 patients included, the mean age and standard deviation (SD) was 54±28.3 years, 67 (64.7%) were male and 35 (35.3%) were female (Table 1). Mean dialysis duration before transplantation and SD was 104±31.6 days. Overall,
92 transplants (88.5%) were from deceased donors whom 40 (38.5%) had expanded criteria. The mean and SD body mass index and skeletal muscle index were, respectively, 25.1±4.2 kg/m2 and 44.8±12.1 cm2/m2. The mean muscle surface and SD were 126±46.2 cm2 and its density was 35.4±11.6 Hounsfield unities.
One‑month post‑transplantation complications At 1 month post-transplantation, 67 patients (63.9%) had a complication whom 14 (13.8%) were considered major, 3 patients (2.9%) had transplant removal and 2 patients (2%) died. The details of these complications are described in Supplementary Table 1. In univariate logistic regression analysis, plasmatic albuminemia, the use of anticoagulation and warm ischemia time was significantly associated with Clavien–Dindo≥3 postoperative complications [respectively OR (95% CI), 0.2 (0.1–0.6), 7.6 (2.4–28.6) and 0.9 (0.8–0.98)] (Table 2). Finally, low plasmatic albuminemia and the use of anticoagulants were risk factors of postoperative complication in multivariable analysis [respectively OR (95%CI), 0.3 (0.1–0.9) p=0.05 and 6.4 (1.8–27.4) p=0.01].
One‑year post‑transplantation surgical complications After a 1-year follow-up, six patients had died (5.9%) and five patients (4.9%) returned to dialysis (Table 3). Overall, 62 patients (60.8%) had a medical complication and 30 (29.4%) a surgical complication in the year following transplantation. In univariate analysis, muscle density and the use of anticoagulation were significantly associated with the occurrence of surgical complication [respectively OR (95% CI), 0.6 (0.4–0.9) and 2.8 (0.9–10.6)] (Table 4). In a multivariate analysis including variables whose p-value ≤ 0.1, a low muscle density and a residual diuresis remained risk factors of 1-year surgical complications [respectively OR (95% CI), 0.6 (0.3–0.9) p=0.05 and 4.9 (1.2–23) p=0.05]. The area under the curve (AUC) of a 1-year complication predictive model including residual diuresis and muscle density was 0.64 (Fig. 2).
Discussion
In our study, we assessed the association between sarcopenia and short-term renal transplantation outcomes using CT-scan sections. We demonstrated that patients with lower albuminemia had significantly more complications at 1-month post-transplantation. At 1 year, a low muscle density was associated with surgical complications when the albuminemia seemed not to have any impact. The use of anticoagulants was also correlated with the complication rate, at 1 month and 1 year after transplantation. The prevalence of sarcopenia in the dialysis population is ranging from 5 to 37%, depending on the stage of renal failure [7, 17]. In this context, protein-energy undernutrition can be explained by several mechanisms: (1) decreased intake, due to restrictive diets; (2) disruption of protein metabolism caused by low physical activity, increased catabolism related to metabolic acidosis, and a decreased anabolism related to peripheral resistance to insulin; (3) protein loss due to proteinuria [18]. Prolonged dialysis thus contributes to the metabolic and nutritional disorder sufered by patients with end-stage renal disease [19]. Although it has been shown that sarcopenia was associated with postoperative outcomes in the field of oncology, its impact in renal transplantation is not clearly established.
Some retrospective studies in lung and liver transplantation highlighted the pejorative prognostic of pre-transplant sarcopenia [20, 21]. Consequently, our study was one of the few to evaluate the impact of CT-scan measured sarcopenia and muscle density on post-transplantation outcomes. Our method provided an objective loss account and muscle tissue alteration, even in individuals with normal or supra-normal BMI [22]. Because dialysis patients have a wide range of weights, we felt that this method was particularly suitable among this population. Moreover, it has already been validated as an efficient tool to assess sarcopenia [9, 23]. We could not evidence a significant association between sarcopenia and complications at 1 month and 1 year after the first kidney transplant. However, the definition of sarcopenia is still not consensual at present. Based on the measurement of muscle surface area on a CT-scan section, there is currently no consensus value for defining sarcopenia. Furthermore, its definition has been revised by the European Work Group on Sarcopenia in Older People (EWGSOP2) in 2018 and it is mainly clinical. Indeed, the co-existence of two of the following three factors now allows sarcopenia diagnosis: a decrease in muscle strength, and/or a decrease in muscle quantity, and/or a decrease in performance on physical tests [10]. This definition is therefore now based more on functional than anatomical criteria. A disturbance in muscle quality thus prevails over the muscle quantity itself. From this perspective, muscle density could be an interesting and early tool to evaluate the nutritional status of patients in the waiting list. However, we could evidence a relation between muscle density and post-transplantation outcomes. Indeed, muscle density decreasing, reflecting the degree of fatty infiltration of muscle tissue, could be the first step in the process of muscle degeneration responsible for function alteration. Muscle density is also comparable in patients regardless of their gender or body mass index ensuring generalization [24].
Our study presents several limitations. First, our study design (monocentric and retrospective) and the low number of subjects could bias our results. Second, half of the patients who underwent renal transplantation during the study period were excluded because they did not have a CT scan in the last 12 months before surgery. At this time, the limitation was chosen to have the most identical morphometric patient profile as on the transplantation day. Moreover, as sarcopenia is likely non-static and likely to change over time, using the 18-month duration exposure could also bias our results. The ideal delay would have been to undergo CT scan the day before renal transplantation among an appropriate prospective study; however, we believe that this study should be seen as a proof of concept for further studies. Finally, CT-scan assessment of sarcopenia is not yet a standard and patients could have been misclassified. In this scope, the evaluation of muscle density seems more objective and could be
Fig. 2 ROC analysis of the multivariable model predicting 1-year surgical complication. The AUC of a 1-year complication predictive model including residual diuresis and muscle density was 0.64 a lead for better post-transplantation outcomes prediction. Further studies on the subject are still needed to establish more clearly the impact of sarcopenia on the occurrence of complications after kidney transplantation.
Conclusion
The occurrence of complications at 1 month and 1 year after a first renal transplant did not seem to be associated with the sarcopenic status of patients. However, CT-scan muscle density and plasmatic albuminemia were associated with pejorative post-transplantation outcomes and could prefigure as early predictive tools for these patients.
