Obstruction caused by ureteral stones, malignancies, or other benign factors can lead to HUN, acute renal failure, and, when accompanied by infection, pyelonephritis. Acute renal failure is rarely observed in healthy individuals unless there is bilateral involvement; however, in patients with comorbidities, the disease progression may be significantly accelerated [7]. Among patients with malignancies, the reactive inflammatory response may be either exaggerated or suppressed compared to that in healthy individuals [8]. This variability can be attributed to tumor-related pathophysiology or to the effects of therapeutic agents, such as chemotherapeutics, hormonal treatments, and immunotherapy.
When sepsis-related findings are analyzed, low platelet levels—along with elevated procalcitonin and CRP levels—emerge as significant markers in this patient group (Table 2). Platelets play a pivotal role in modulating the immune response during infection and maintaining vascular integrity by supporting endothelial function. In the context of sepsis, increased platelet consumption and destruction contribute to thrombocytopenia. This excessive consumption and activation of platelets may further amplify the dysregulated immune response, potentially leading to coagulopathy, multiorgan dysfunction, and increased mortality. Notably, when platelet counts fall to ≤50,000/μL, the risk of organ failure and mortality increases significantly [9–12]. Similarly, procalcitonin and CRP as essential biomarkers for assessing the severity of infection and evaluating the immune response. These markers have been consistently validated as reliable indicators across numerous studies [11]. Even though leukocytosis is commonly observed in sepsis, leukopenia may also occur, as highlighted in the SIRS criteria. In our patient cohort, neutropenia and lymphopenia were observed at statistically significant levels among those who developed sepsis (Table 2,3). The severity of neutropenia was further supported by a decrease in post-op NLR compared to pre-op values. NLR is considered a reliable marker that reflects both the intensity of the inflammatory response and the functional state of the immune system. Neutrophils constitute the first line of defense against microorganisms during infection, whereas lymphocytes reflect the activity of the adaptive immune response. An increase in neutrophil counts accompanied by a decrease in lymphocyte counts during sepsis indicates a dysregulated immune response and the uncontrolled progression of inflammation [8,9,12–14]. Interestingly, although the existing literature generally reports an elevated NLR during inflammatory conditions, our study found that NLR was lower in the sepsis group, which consisted exclusively of patients with malignancies (Table 2–3). We believe this situation is attributable to immune system dysfunction resulting from the coexistence of malignancy and sepsis [15,16]. The fact that post-op NLR emerged as an independent predictor of mortality in both univariate and multivariable analyses further highlights the prognostic value of this parameter.
In our study, ICU admission emerged as an independent predictor of sepsis in both univariate and multivariable analyses. This finding indicates that patients requiring ICU-level care are at significantly higher risk for developing sepsis, even after adjusting for other clinical factors. ICU admission likely reflects early physiological deterioration and increased disease burden. Therefore, the presence of ICU-level needs should be regarded as an early warning sign, prompting heightened clinical vigilance and timely interventions to prevent or mitigate sepsis. Another independent predictor of sepsis identified in our study was the presence of perirenal fat stranding (Table 2–3). Inflammation within the perirenal fat tissue contributes to an increased microbial burden and a heightened risk of bacterial infections. It is well known that invasive interventions under such inflammatory conditions can significantly elevate the risk of sepsis [5,17]. The Mayo Adhesive Probability (MAP) score is a valuable tool for evaluating the inflammatory burden of perirenal fat tissue and estimating the associated risks of infection and surgical complications. An elevated MAP score serves as a reliable indicator of severe perirenal inflammation and the potential progression of infection [18–20]. The integration of predictive scoring systems, such as the MAP score, into the clinical management of similar patient populations may enhance clinical decision-making and optimize treatment strategies.
When evaluating the parameters associated with mortality, it becomes evident that, despite pathophysiological similarities with sepsis, certain distinctions are observed. Notably, neutropenia and thrombocytopenia do not emerge as significant predictive factors (Table 2,4). In contrast, low pre-op and post-op lymphocyte levels, reduced post-op NLR, elevated pre-op and post-op procalcitonin and post-op CRP levels appear as key predictors of mortality, consistent with their roles in sepsis. However, among these variables, only post-op NLR was found to be an independent predictor of mortality after nephrostomy in logistic regression analysis (Table 2,4). Lymphopenia reflects a weakened immune system in the context of infection and indicates an impaired adaptive immune response [16,21,22]. Additionally, creatinine levels following nephrostomy were identified as predictors of mortality in univariate analysis; however, this association did not remain significant in multivariable analysis (Table 2,4). Although mortality rates were statistically higher among patients with non-urological malignancies, logistic regression analysis revealed no significant difference in mortality between patients with urological and non-urological malignancies (Table 2,4). Diabetes mellitus and ISDU complicate management of infections, particularly in the context of immunosuppression induced by sepsis and malignancy. These conditions exacerbate septic progression, increase the likelihood of complications, and elevate the risk of mortality [15,23,24]. Interestingly, while a statistically significant association (p<0,05) was found between DM and ISDU with mortality in our cohort, logistic regression analysis showed that they were not independent predictors of mortality. These results suggest that although DM and ISDU may indicate disease severity, they do not necessarily translate into increased mortality risk in all settings, highlighting the need for individualized patient assessment (Table 2,4). ICU admission emerged as a key predictor of both sepsis and mortality. ICU management strategies play a crucial role in directly influencing patient outcomes. The literature emphasizes that early ICU admission and timely implementation of supportive therapies can significantly reduce mortality rates [25]. In this context, the strong association between ICU admission and mortality observed in our study is consistent with findings in the literature. Furthermore, among the parameters variables included in multivariable logistic regression analysis, post-op ICU admission was identified as independent predictor of mortality (Table 4).
This study has several limitations, including its retrospective design, relatively small sample size, and single-center setting. Additionally, more detailed classification of malignancy types could have better illustrated the diversity of etiopathogenesis and inflammatory responses. The inclusion of data regarding chemotherapeutic agents, hormonal therapies or immunotherapies might have helped reduce heterogeneity and allowed for a more nuanced analysis. Due to the retrospective nature of our study, it was not possible to clearly differentiate between primary causes of mortality, which may have contributed to the high mortality rate observed. We believe that the actual mortality rate is likely lower, as our cohort consisted exclusively of cancer patients, and cancer-related deaths may have inflated the observed rate—constituting another important limitation of this study. Nevertheless, this study provides a foundation for future randomized prospective studies with larger cohorts, to evaluate dynamic changes in inflammatory markers. Including patients with benign causes of nephrostomy as a control group would enhance the understanding of malignancy-specific outcomes. Furthermore, assessing long-term results and quality of life would contribute the development of more comprehensive and evidence-based guidelines for the management of patients requiring nephrostomy.
DISCUSSION
Obstruction caused by ureteral stones, malignancies, or other benign factors can lead to HUN, acute renal failure, and, when accompanied by infection, pyelonephritis. Acute renal failure is rarely observed in healthy individuals unless there is bilateral involvement; however, in patients with comorbidities, the disease progression may be significantly accelerated [7]. Among patients with malignancies, the reactive inflammatory response may be either exaggerated or suppressed compared to that in healthy individuals [8]. This variability can be attributed to tumor-related pathophysiology or to the effects of therapeutic agents, such as chemotherapeutics, hormonal treatments, and immunotherapy.
When sepsis-related findings are analyzed, low platelet levels—along with elevated procalcitonin and CRP levels—emerge as significant markers in this patient group (Table 2). Platelets play a pivotal role in modulating the immune response during infection and maintaining vascular integrity by supporting endothelial function. In the context of sepsis, increased platelet consumption and destruction contribute to thrombocytopenia. This excessive consumption and activation of platelets may further amplify the dysregulated immune response, potentially leading to coagulopathy, multiorgan dysfunction, and increased mortality. Notably, when platelet counts fall to ≤50,000/μL, the risk of organ failure and mortality increases significantly [9–12]. Similarly, procalcitonin and CRP as essential biomarkers for assessing the severity of infection and evaluating the immune response. These markers have been consistently validated as reliable indicators across numerous studies [11]. Even though leukocytosis is commonly observed in sepsis, leukopenia may also occur, as highlighted in the SIRS criteria. In our patient cohort, neutropenia and lymphopenia were observed at statistically significant levels among those who developed sepsis (Table 2,3). The severity of neutropenia was further supported by a decrease in post-op NLR compared to pre-op values. NLR is considered a reliable marker that reflects both the intensity of the inflammatory response and the functional state of the immune system. Neutrophils constitute the first line of defense against microorganisms during infection, whereas lymphocytes reflect the activity of the adaptive immune response. An increase in neutrophil counts accompanied by a decrease in lymphocyte counts during sepsis indicates a dysregulated immune response and the uncontrolled progression of inflammation [8,9,12–14]. Interestingly, although the existing literature generally reports an elevated NLR during inflammatory conditions, our study found that NLR was lower in the sepsis group, which consisted exclusively of patients with malignancies (Table 2–3). We believe this situation is attributable to immune system dysfunction resulting from the coexistence of malignancy and sepsis [15,16]. The fact that post-op NLR emerged as an independent predictor of mortality in both univariate and multivariable analyses further highlights the prognostic value of this parameter.
In our study, ICU admission emerged as an independent predictor of sepsis in both univariate and multivariable analyses. This finding indicates that patients requiring ICU-level care are at significantly higher risk for developing sepsis, even after adjusting for other clinical factors. ICU admission likely reflects early physiological deterioration and increased disease burden. Therefore, the presence of ICU-level needs should be regarded as an early warning sign, prompting heightened clinical vigilance and timely interventions to prevent or mitigate sepsis. Another independent predictor of sepsis identified in our study was the presence of perirenal fat stranding (Table 2–3). Inflammation within the perirenal fat tissue contributes to an increased microbial burden and a heightened risk of bacterial infections. It is well known that invasive interventions under such inflammatory conditions can significantly elevate the risk of sepsis [5,17]. The Mayo Adhesive Probability (MAP) score is a valuable tool for evaluating the inflammatory burden of perirenal fat tissue and estimating the associated risks of infection and surgical complications. An elevated MAP score serves as a reliable indicator of severe perirenal inflammation and the potential progression of infection [18–20]. The integration of predictive scoring systems, such as the MAP score, into the clinical management of similar patient populations may enhance clinical decision-making and optimize treatment strategies.
When evaluating the parameters associated with mortality, it becomes evident that, despite pathophysiological similarities with sepsis, certain distinctions are observed. Notably, neutropenia and thrombocytopenia do not emerge as significant predictive factors (Table 2,4). In contrast, low pre-op and post-op lymphocyte levels, reduced post-op NLR, elevated pre-op and post-op procalcitonin and post-op CRP levels appear as key predictors of mortality, consistent with their roles in sepsis. However, among these variables, only post-op NLR was found to be an independent predictor of mortality after nephrostomy in logistic regression analysis (Table 2,4). Lymphopenia reflects a weakened immune system in the context of infection and indicates an impaired adaptive immune response [16,21,22]. Additionally, creatinine levels following nephrostomy were identified as predictors of mortality in univariate analysis; however, this association did not remain significant in multivariable analysis (Table 2,4). Although mortality rates were statistically higher among patients with non-urological malignancies, logistic regression analysis revealed no significant difference in mortality between patients with urological and non-urological malignancies (Table 2,4). Diabetes mellitus and ISDU complicate management of infections, particularly in the context of immunosuppression induced by sepsis and malignancy. These conditions exacerbate septic progression, increase the likelihood of complications, and elevate the risk of mortality [15,23,24]. Interestingly, while a statistically significant association (p<0,05) was found between DM and ISDU with mortality in our cohort, logistic regression analysis showed that they were not independent predictors of mortality. These results suggest that although DM and ISDU may indicate disease severity, they do not necessarily translate into increased mortality risk in all settings, highlighting the need for individualized patient assessment (Table 2,4). ICU admission emerged as a key predictor of both sepsis and mortality. ICU management strategies play a crucial role in directly influencing patient outcomes. The literature emphasizes that early ICU admission and timely implementation of supportive therapies can significantly reduce mortality rates [25]. In this context, the strong association between ICU admission and mortality observed in our study is consistent with findings in the literature. Furthermore, among the parameters variables included in multivariable logistic regression analysis, post-op ICU admission was identified as independent predictor of mortality (Table 4).
This study has several limitations, including its retrospective design, relatively small sample size, and single-center setting. Additionally, more detailed classification of malignancy types could have better illustrated the diversity of etiopathogenesis and inflammatory responses. The inclusion of data regarding chemotherapeutic agents, hormonal therapies or immunotherapies might have helped reduce heterogeneity and allowed for a more nuanced analysis. Due to the retrospective nature of our study, it was not possible to clearly differentiate between primary causes of mortality, which may have contributed to the high mortality rate observed. We believe that the actual mortality rate is likely lower, as our cohort consisted exclusively of cancer patients, and cancer-related deaths may have inflated the observed rate—constituting another important limitation of this study. Nevertheless, this study provides a foundation for future randomized prospective studies with larger cohorts, to evaluate dynamic changes in inflammatory markers. Including patients with benign causes of nephrostomy as a control group would enhance the understanding of malignancy-specific outcomes. Furthermore, assessing long-term results and quality of life would contribute the development of more comprehensive and evidence-based guidelines for the management of patients requiring nephrostomy.