Remember me
Click on image for details.
EDITORIAL COMMENTARY
Year : 2023 | Volume
: 69
| Issue : 4 | Page : 192-193
Post-transplant diarrhea and immunosuppressant withdrawal: Insights and limitations
M Kulkarni
Correspondence Address:
Dr. M Kulkarni
Department of Nephrology, Father Muller Medical College, Mangalore, Karnataka
India
Source of Support: None, Conflict of Interest: None
CheckDOI: 10.4103/jpgm.jpgm_462_23
How to cite this article:Innocent, unbiased observation is a myth.
-Medawar, Sir Peter Brian (1915–87)[1]
Kidney transplantation (KTR) is the preferred treatment for patients with end-stage renal disease, offering good quality of life. However, medical complications are frequently observed following KTR. Among gastrointestinal complications that occur after KTR, diarrhea is particularly common,[2] with an incidence ranging from 13 to 64%. According to US Renal Data System (USRDS), unspecified noninfectious diarrhea is associated with an increased risk of graft loss and patient death.[3]
In this issue of the journal, Patil et al.[4] present observational data of 85 KTR recipients who experienced diarrhea. This retrospective analytic study focuses on 41 patients who developed chronic allograft injury (CAI) following post-transplant diarrhea (PTD). The descriptive part of the study provides valuable insights into the stepwise evaluation of patients with PTD and identifies the etiology of diarrhea, thereby contributing important data specific to this geographical region. Approximately 65% of the episodes were found to be infective, and conventional stool examination successfully diagnosed more than one-third of the cases. This observation provides confidence to clinicians that a simple stool examination can help diagnose one-third of PTD cases. Additionally, more than 90% of PTD patients developed acute kidney injury (AKI), with 75% of them recovering to their baseline renal function.
When interpreting the results of the analytic part of the study, caution is required, particularly regarding the observation of CAI in nearly half of the PTD cases. The authors note an association between diarrhea and CAI, which may be confounded by various factors such as the etiology and duration of diarrhea, as well as the use of immunosuppressants. One of the most significant confounders is the modification of immunosuppression, specifically the withdrawal or dose reduction of mycophenolate mofetil (MMF). In this study, MMF was completely stopped in 35 out of 73 patients. It becomes challenging to quantify the role of MMF withdrawal in CAI as opposed to PTD.
The authors employ a log-rank test to support their observation that MMF withdrawal is associated with CAI (HR 0.46, 95% CI, 0.22–0.94). However, a Chi-square test performed on the same data fails to find any association. These discrepancies are inherent problems with observational data. The use of Kaplan-Meier curves and log-rank tests has limitations when working with retrospective data due to inaccuracies in determining the exact time of the event and obtaining data on patients lost to follow-up. Another drawback of using survival analysis will be censoring. Some patients might not develop the outcome of interest (CAI) in the defined observation time frame, some might have died and some might simply have stopped following up at the center. This rightward and leftward censoring can limit the interpretation of results, more so in retrospective dataset than in randomized trials. Therefore, caution is necessary when interpreting the association between MMF withdrawal and CAI.
There are additional limitations that restrict the generalizability of the results. Firstly, the study population consists of kidney transplant recipients who sought medical attention for diarrhea. The definition of diarrhea can vary, which poses a challenge when working with retrospective data. Furthermore, the study was conducted at a tertiary center, introducing the possibility of Berksonian bias.[5] The guidelines for MMF withdrawal after diarrhea are not well-established and likely to vary across different centers, further limiting the generalizability of the results. Another significant concern is the choice of outcome, namely CAI, and its definition. The study defines CAI as a creatinine level of ≥1.5 mg/dl persisting for ≥3 months, with an increase of 25% from baseline in recipients with abnormal eGFR. Using such a sensitive definition of CAI may have led to an overestimation of its impact and incidence. Moreover, CAI defined in this manner is not a reliable surrogate for clinical outcomes.[6] Studying graft loss and death would have been more meaningful outcomes to investigate.[7]
In addition to the aforementioned limitations, the study sheds light on challenges commonly encountered in observational studies.[8] Most statistical tests in these studies are conducted after data collection, increasing the likelihood of obtaining false positive results.
Despite these limitations, the study offers valuable observational data on PTD. The potential association between diarrhea, MMF withdrawal, and CAI is an important finding that requires careful interpretation. Future studies should place emphasis on key outcomes such as graft failure following diarrhea and MMF withdrawal while considering other contributing risk factors.
:: References
Comments (0)