Patient profiles were created using demographic characteristics, laboratory data, comorbidities, and medication-related information collected before and during hospitalization. MRPs were identified through systematic medication review, medication reconciliation at admission, therapeutic drug monitoring, and assessment of drug selection and dosing. Routine clinical pharmacy services, including patient counseling and medication education, were provided as part of standard clinical care.

Pediatric patients were defined as those who were less than 18 years of age. Polypharmacy was defined as the concomitant use of five or medications (Masnoon et al., 2017). Patients estimated glomerular filtration rates were calculated using the 2021 Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) formula (Lee et al., 2025). Chronic kidney disease (CKD) was defined as structural or functional kidney abnormalities persisting for at least 3 months (Stevens et al., 2024). CKD stages, acute kidney injury (AKI), and renal replacement therapy were all recorded. Drug selection, dose regimens, and potential drug-drug interactions were evaluated based on national and international clinical guidelines using UpToDate ® and Lexicomp ®. All identified MRPs and related recommendations were communicated to the responsible physician or healthcare staff. MRPs, interventions, and outcomes were classified using the PCNE v9.1 system (Van Mil et al., 2020).

The statistical analysis of the data was performed using IBM SPSS Statistics (Statistical Package for the Social Sciences) version 27.0. The study was designed as a prospective observational inquiry to include all eligible liver transplant patients admitted within a specified timeframe, rather than as a hypothesis-driven study with a preset sample size. Thus, the study population consisted of all patients who met the inclusion criteria during the study period, forming a consecutive cohort representative of routine clinical practice. The normality of quantitative data was assessed using the Kolmogorov-Smirnov test. As the data did not conform to a normal distribution, continuous variables were summarized as medians. Quantitative data from two groups were compared using the Mann-Whitney U test, while comparisons across more than two groups were conducted using the Kruskal-Wallis H test. The chi-square test was applied for comparisons of categorical variables, and relationships between qualitative variables in two groups were also assessed using the chi-square test. Multivariate analysis was conducted using logistic regression. Variables included in the multivariable logistic regression model were those showing a statistically significant association at the level by 0.2 with MRPs in univariable analyses (p < 0.2). Given the limited number of events, a highly parsimonious model was intentionally constructed to minimize overfitting and model instability. The variance explained by the model was indicated by the Nagelkerke R² value, and model fit was evaluated using the Hosmer-Lemeshow test. Risk factors were presented as odds ratios (OR). A p-value of <0.05 was considered statistically significant.

In this study, 62.2% of the included patients were male. In the literature, similar gender distributions have been reported, with male rates of 60.9% by Murder et al., 75.9% by Pinheiro et al., and 57.8% by Shawaqfeh et al. (Mulder et al., 2021; Pinheiro et al., 2020; Shawaqfeh et al., 2023). These findings align with the literature, indicating that liver transplantation is more commonly performed in male patients. The median age of all patients in this study was 50 years (IQR: 35–61). Regarding the ages of liver transplant recipients, Schuh et al. reported a median age of 59.7 years in their study, while Mulder et al. found a median age of 59.5 years (IQR: 47–66) (Yang et al., 2019; Mulder et al., 2022). These international studies reported consistent median age values among liver transplant recipients. The comparatively lower median age in this cohort is likely attributable to the inclusion of pediatric patients who were hospitalized in the ward or ICU. When evaluating the relationship between patients’ demographic characteristics and the incidence of medication-related problems (MRPs), no significant relationship was found between gender or age and MRP occurrence. Similar results were also reported in the literature (Mulder et al., 2021; Martin and Zavala, 2004; Azadi et al., 2022).

The median number of medications used by patients in this study was 15 (IQR: 12–19; range: 8–30). In a study carried out on solid organ transplant (SOT) recipients, the mean number of medications used was reported to be 10.9 ± 3.9 (20). Another study investigating the effects of polypharmacy on transplant recipients revealed that, in the early post-transplant period, an average of 8 medications were prescribed, with the number increasing to 10–15 due to chronic comorbidities (Kim et al., 2023). These results are consistent with the result of present study. Furthermore, the total number of medications was significantly higher in the group with MRPs (p < 0.05). Similarly, O’Grady et al. reported that polypharmacy may reduce medication adherence and contribute to the development of MRPs (O'Grady et al., 2010). This may be due to the complex and demanding therapeutic regimens that transplant recipients must adhere to throughout their lives to treat or prevent complications.

The most frequently used drug classes among the study population were immunosuppressive agents (100%), proton pump inhibitors (99.5%), and antimicrobial agents (99.5%). In a study conducted by Duwez et al., the most commonly used medications were also immunosuppressants, followed by antimicrobials (Duwez et al., 2020). Consistent with the results achieved in the present study, Yang et al. reported that the most frequently used drug classes were immunosuppressive, antimicrobial, and antihypertensive agents, respectively (Yang et al., 2019).

The most common comorbidities among the patients in this study were hypertension (45.3%), diabetes mellitus (39.9%), chronic kidney disease (35.9%), and cardiovascular diseases (26.3%). Even though Mulder et al. reported slightly different rates, the most prevalent comorbidities were also hypertension (59%), diabetes (26%), and renal diseases (11.3%) (Mulder et al., 2022). Another study conducted in 2021 revealed the most frequently observed comorbidities to be chronic kidney disease (40.6%) and diabetes (29.7%) (Mulder et al., 2021).

In this study, the presence of at least one comorbidity increased the risk of MRPs by 3.3 times. In contrast, Azadi et al. found that having more than three comorbidities increased the risk of MRPs by 1.65 times (Azadi et al., 2022). The discrepancy in risk ratios may be due to differing thresholds used to define comorbidity-related risk. Interestingly, when comorbidities were examined, patients with a history of diabetes showed a decreased risk of MRPs. This is thought to be a coincidental finding associated with the small number of patients without MRPs in this group. On the other hand, the presence of acute kidney injury increased the risk of MRPs by 15.16 times. A literature review revealed that there was no study individually assessing comorbidities as risk factors for MRPs in transplant patients. The observed association between MRPs and acute kidney injury likely reflects overall disease severity and clinical complexity rather than a direct causal relationship.

The higher MRP burden observed among ICU patients and those with renal dysfunction suggests the presence of confounding by indication, as greater disease severity necessitates more complex pharmacotherapy and is independently associated with adverse outcomes.

In the literature, the number of studies in the literature evaluating MRPs in solid organ transplantation (SOT), including liver transplantation, is limited.

In this study, at least one MRP was identified in 311 patients (83.37%) who were admitted to either the ICU or inpatient wards. Consistent with these findings, Mulder et al. reported MRPs in 87.8% of patients, while Bonkowski et al. detected MRPs in 83.8% of SOT recipients (Mulder et al., 2022; Bonkowski et al., 2014). Although the reported incidence rates vary slightly across studies, the literature consistently demonstrates a high prevalence of MRPs in liver transplant recipients.

In this study, the most frequently observed type of MRP fell under the category of “treatment effectiveness” (P1), accounting for 59.8% of all problems, followed by “treatment safety” (P2), at 34.4%. Similarly, in a study that evaluated MRPs in SOT recipients monitored in the ICU due to COVID-19 using the PCNE classification system, “treatment effectiveness” was also the most common MRP category, comprising 34.12% of all problems, followed by “treatment safety” at 33.02% (Azadi et al., 2022). Since the MRPs in other studies involving similar patient populations were not classified according to the PCNE system, direct comparisons by problem type are not feasible. A review of the literature indicates that this study is the first and most comprehensive to classify MRPs using the PCNE v9.1 system in liver transplant recipients hospitalized in both the ICU and inpatient wards.

When focusing solely on liver transplant recipients in the ICU, the most frequently observed MRPs were again related with “treatment effectiveness” (P1), accounting for 38.5%, followed by “treatment safety” (P2), at 23.3%. In line with the findings achieved in this study, Li et al. also reported that “treatment effectiveness” (P1) was the most common category, at 41%, in ICU patients (Li et al., 2020). In other studies, evaluating MRPs using the PCNE v9.1 system, issues classified under “treatment effectiveness” (P1) became the most common category, accounting for 47.9% and 50.4% of all MRPs (Kara et al., 2024; Çakır et al., 2024). However, a study utilizing PCNE v9.0 in ICU patients reported that 43.4% of MRPs were related with “treatment safety” (P2), and another similar study found this rate to be as high as 77.18% (Durmuş et al., 2024; Albayrak et al., 2022). These variations suggest that differences in the most frequently encountered MRP categories may be attributable to differences in patient populations across studies.

A review of the literature reveals no existing studies investigating MRPs associated with renal dose adjustment in liver transplant recipients. In this study, among ICU patients with renal impairment who required renal dose adjustments, 24.3% had at least one MRP related to these medications. Notably, these problems accounted for the majority of those categorized under “treatment safety” (P2). In a study carried out by Garin et al., the incidence of MRPs associated with renal dose adjustments in ICU patients was reported as 17%, while Alsayed et al. reported a rate of 22% (Garin et al., 2021; Ali et al., 2024).

In this study, the most frequently encountered cause of MRPs was “dose selection” (C3), accounting for 44.2% of cases, followed by “drug selection” (C1) at 26.1%. Similarly, in a study carried out by Repp et al. among heart transplant patients, “dose selection” was reported as the most common cause of MRPs at 48.2%. This was followed by Duwez et al.’s study in lung transplant recipients (39.6%) and Lee et al.’s study in kidney transplant recipients (38%) (Bonkowski et al., 2014; Lima et al., 2016; Lee et al., 2016). These findings suggest that clinical pharmacists’ recommendations regarding dose selection can significantly contribute to the prevention of MRPs.

When evaluating MRP causes identified exclusively in critically ill patients, the most frequent issue was again found to be related with “dose selection” (C3), accounting for 30% of cases, consistent with the overall findings of this study. In a study carried out by Kara et al., “dose selection” was also the most frequently observed MRP cause, reported at 77% (Kara et al., 2024).

Within the category of MRPs classified under “drug selection,” the most common subcategory was “absence of drug therapy despite an existing indication” (C1.5), which accounted for 17.8% of cases. This aligns with the findings achieved by Mulder et al., who reported an untreated indication rate of 15%, and those achieved by Wang et al., who found this rate to be 20.9% in a study on kidney transplant recipients (Mulder et al., 2022; Wang et al., 2008).

In this study, the most frequently observed subcategory under the “other” classification was “treatment outcome not monitored or monitored inappropriately (including therapeutic drug monitoring)” (C9.1), at 14.4%. In line with these results, Duwez et al. reported a frequency of 11.2% for this issue, Covert et al. reported 13% in kidney transplant patients, and Yang et al. found a rate of 10% in their study on kidney transplant recipients (Duwez et al., 2020; Yang et al., 2019; Covert et al., 2017).

A total of 620 suggestions were offered for the 620 identified MRPs in this study. The majority of these suggestions were directed at the “drug level” (I3) (43.45%) and at the “prescribing physician level” (I1) (41.2%).

Among the suggestions related with MRPs at the “drug level” (I3), the most common suggestion was “dose changed to…” (I3.2), which accounted for 18% of the suggestions. Consistent with these findings, Pinheiro et al. reported a similar rate of 18.2% for dose change suggestions, Wang et al. reported 15%, and Lima et al. found a rate of 10.2% (Pinheiro et al., 2020; Wang et al., 2008; Lima et al., 2016). In a study carried out by Duwez et al., change recommendations constituted 43.8% of all interventions, while this proportion was 33.8% in the study carried out by Mulder et al. (Bonkowski et al., 2014; Kara et al., 2024). These variations in reported frequencies are likely attributable to differences in the methods used for MRP classification.

Considering the “drug level” (I3) category, the second most common suggestion was “initiation of a new medication” (I3.6), accounting for 11.2% of the suggestions. In line with these results, Mulder et al. reported a rate of 10.2% for similar suggestions, and Pinheiro et al. reported 11% (Mulder et al., 2021). However, in the study carried out by Azadi et al., “initiation of a new medication” (I3.6) was reported at a significantly lower rate of 1.83% (Azadi et al., 2022). The discrepancy among studies may be due to additional suggestions related to patients’ therapeutic regimens during the COVID-19 pandemic.

Of the suggestions offered for MRPs identified in this study, 88.2% were accepted. This acceptance rate is comparable to those reported in the literature, such as 93% reported by Lee et al. for kidney transplant recipients, 93.6% by Mulder et al., and 95% in another study involving both kidney and liver transplant recipients (Pinheiro et al., 2020; Mulder et al., 2022; Lee et al., 2016). These findings highlight the high acceptance rates of clinical pharmacists’ suggestions worldwide in promoting safe and effective pharmacotherapy.

Although the PCNE v9.1 classification was employed as a standardized framework for the systematic categorization of MRPs, its standalone presentation may be perceived as largely descriptive if not accompanied by clinical context. To enhance clinical interpretability, a Supplementary Table S1 has been added, providing representative, real-word clinical examples from our cohort. This table illustrates the application of the PCNE v9.1 classification across the problem, cause, intervention and outcome domains, thereby contextualizing the classification within routine clinical practice.

The ADEs observed in the present study illustrate key pharmacotherapy-related safety risks inherent to liver transplantation. Predominant renal, hematological, neurological, and electrolyte disturbances associated with commonly used immunosuppressive and antimicrobial agents highlight the narrow therapeutic margins and cumulative toxicity burden in this population. These findings provide clinically relevant safety context for MRPs and underscore the importance of systematic medication review and close pharmacotherapy monitoring in routine transplant care.

In the evaluation of MRPs, particularly in highly complex patient populations, a delicate balance exists between the standardized recommendations provided by clinical practice guidelines and individualized, patient-specific clinical decision-making processes Although clinical practice guidelines aim to standardize medical care based on scientific evidence, treatment decisions are often influenced by the patient’s clinical condition, comorbidities and broader contextual factors (Mercuri et al., 2015). In this context, some of the MRPs identified in the present study may reflect deliberate deviations from guideline recommendations driven by individualized clinical judgment in specific clinical scenarios, rather than in appropriate treatment practices per se. From a clinical pharmacist perspective, such deviations were classified as MRPs due to their potential risk profile. Moreover, the non-acceptance of certain pharmacist recommendations by the clinical team may represent patient-spesific,context-driven decision-making rather than erroneous clinical practice. Taken together, these conditions suggest that the concept of MRPs should not be viewed solely as indicators of absolute inappropriateness, but rather within a broader framework that also encompasses clinical risk awareness and the need for multidisciplinary evaluation.

Among the strengths of this study are its prospective design, which allowed for direct communication with both patients and healthcare professionals, and its inclusion of the entire transplantation unit. To the best of our knowledge, this is the first prospective study in which all liver transplant recipients admitted to both the ICU and inpatient ward were included, medication-related problems (MRPs) were identified by a clinical pharmacist, suggestions were offered, and all findings were classified using the PCNE v9.1 system.

This study has several limitations. First, its single-center design may limit the generalizability of the findings, and the absence of a control group precludes causal inferences regarding the impact of clinical pharmacy interventions. In addition, no structured educational training program was provided to patients or healthcare staff as part of the study. Although MRPs and pharmacist interventions were prospectively identified and classified, the clinical severity and potential harm associated with individual MRPs were not systematically assessed. Furthermore, patient-centered clinical outcomes following pharmacist recommendations were not evaluated in a structured manner. As the primary aim of the study was to characterize MRPs and pharmacist interventions within a highly complex transplant population, outcome-oriented analyses were beyond the scope of the present investigation.

No formal a priori sample size or power calculation was performed, as the study was designed as a prospective observational investigation including all eligible liver transplant recipients admitted during the predefined study period. Given the limited sample size, subgroup analyses based on key transplant-related factors-such as immunosuppressive burden, renal function, and infection status were not performed, as could have result in small subgroup sizes and increased the risk of model overfitting, thereby compromising the robustness of effect estimates. In addition, despite the use of a conservative variable selection strategy to construct a parsimonious multivariable model, the limited number of events for certain predictors particularly AKI – may have contributed to wide confidence intervals and potential sparse-data bias, warranting cautious interpretation of these findings. Collectively, these limitations may restrict the clinical interpretability of the findings.

Future research should focus on outcome-oriented study designs incorporating validated severity grading of MRPs, structured assessment of patient-centered clinical outcomes (e.g., renal function, infection rates, or length of hospital stay), and appropriate control groups. In addition, adequately powered multicenter studies enabling stratified and subgroup analyses based on key transplant – related factors may further clarify the clinical impact and added value of pharmacist-led interventions in transplant care.

The literature review suggests that this study is the first and most comprehensive investigation to classify MRPs in liver transplant recipients hospitalized in both the ICU and the inpatient ward using the PCNE v9.1 methodology. Furthermore, it is the first study to demonstrate differences in MRP categories between patients in the ICU and those in the general ward. The most common MRP categories identified in both ICU and ward patients were “treatment effectiveness,” with the primary causes being “inappropriate dose selection” and “inappropriate drug selection.

This study is that MRPs in liver transplant recipients frequently reflect the complexity of individualized clinical decision-making rather than isolated prescribing errors. By identifying and contextualizing these risks, clinical pharmacists support rational pharmacotherapy and informed multidisciplinary decision-making in complex transplant care.

The presence of at least one comorbidity and acute kidney injury was found to be an independent risk factors for MRPs in liver transplant recipients, contributing novel data to the existing body of knowledge.

This study concludes that clinicians should exercise particular caution when prescribing new medications to liver transplant recipients with comorbid conditions and a history of acute kidney injury.

Both clinical pharmacists and clinicians should pay close attention to “dose selection” and “drug selection” when reviewing the treatment plans of patients in liver transplant units.

Moreover, this study demonstrates that the active involvement of clinical pharmacists as part of the multidisciplinary team significantly contributes to the successful management of MRPs in transplant recipients with complex medication regimens.