Modified National Early Warning Score 2, warning system for predicting treatment outcomes in patients with emphysematous pyelonephritis

Abstract

Emphysematous pyelonephritis (EPN) is a severe, a lethal necrotizing upper urinary tract infection, characterized by gas production within the renal parenchyma, collecting system, or perinephric tissue. EPN is emerging as a significant concern, necessitating early diagnosis, severity assessment, and timely intervention to improve outcomes. This study proposes a modified National Early Warning Score 2 (mNEWS 2) to enhance risk stratification and predictive accuracy in EPN management. The mNEWS 2 refines the original NEWS 2 system, which aggregates 6 physiological indicators (body temperature, systolic blood pressure, pulse rate, oxygen saturation, breathing rate, and degree of consciousness), by incorporating weighted risk stratification indices and specific cutoff values derived from clinical observations, statistical modeling, and predictive performance analysis. A pilot study identified optimal thresholds, with a score of 15 maximizing predictive performance for mortality risk and intervention needs, validated through receiver operating characteristic curve analysis. So, the mNEWS 2 score represents a significant advancement in EPN management, offering improved risk stratification and treatment outcomes.

Key Words: Modified National Early Warning Score 2; Emphysematous pyelonephritis; Warning system; Pyelonephritis; Emphysematous

Core Tip: Emphysematous pyelonephritis (EPN) is a life-threatening necrotizing infection of the upper urinary tract, predominantly affecting individuals with diabetes mellitus. Early diagnosis and severity assessment are critical due to its high mortality rate. This study proposes a modified National Early Warning Score 2 (mNEWS 2) to enhance risk stratification and predictive accuracy in EPN management. By refining physiological parameter thresholds and incorporating weighted risk indices, mNEWS 2 improves triaging, prioritizes intensive monitoring, and facilitates timely interventions. Validated through receiver operating characteristic curve analysis, the mNEWS 2 score, particularly at a cutoff of 15, maximizes predictive performance for mortality risk and intervention needs.

TO THE EDITOR

The renal parenchyma and, in certain cases, the kidney’s perirenal tissues are affected by emphysematous pyelonephritis (EPN), a severe necrotising infection that affects the upper urinary system. Gas generation in the kidney parenchyma, collecting system, or perinephric tissue is a common symptom of this condition.

Diabetes mellitus is the most common risk factor, affecting more than 90% of people diagnosed with EPN[1]. EPN is a deadly condition, with documented fatality rates ranging from 40% to 90%. The bedside detection of the illness is similar to acute pyelonephritis, which necessitates appropriate imaging, particularly computed tomography (CT) scans. EPN treatment has evolved from severe surgical intervention to more conservative therapeutic techniques, primarily including percutaneous drainage and antimicrobials[2].

Diabetes mellitus is on the rise worldwide, and because it is a risk factor for EPN, it is considered an emerging concern.

Because it is a life-threatening problem with a wide range of treatment options, early diagnosis, and severity assessment are critical.

CLINICAL BACKGROUND

EPN infection can cause nonspecific symptoms, the most common being fever, flank pain, and nausea. Progression to sepsis, multiorgan dysfunction syndrome, and septic shock can be very rapid, especially in uncontrolled diabetic patients and immunocompromised patients with a high mortality index.

Three distinct schemas can be used to categorise EPN according to radiological findings: Michaeli et al[3] initially categorised EPN in 1984 using the results of an intravenous pyelogram and a plain abdomen film of the kidney, ureter, and bladder, Wan et al[4] devised a classification system in 1996 that divided patients into two categories according to computed tomography (CT) results, Huang and Tseng proposed an alternative categorisation in 2000 that included took into account the degree of illness seen on CT scans[5].

So, relying on clinical classification is more accurate and provides an early, fast, and accessible system for illness severity stratification, allowing for early appropriate therapy and lowering mortality.

By improving risk categorisation, this study suggested altering the National Early Warning Score 2 (mNEWS 2) to increase its predictive ability and enable more accurate triage and early intervention[6]. In order to improve decision-making and patient survival in EPN, mNEWS 2 attempts to close the gap between early warning and outcome prediction by customising the score to better represent clinical severity and responsiveness to treatment. Additionally, by putting forward a mNEWS 2 score that improves risk differentiation, permits more accurate triage, prioritises intense monitoring, and enables prompt interventions in EPN cases, this work fills this knowledge vacuum.

A straightforward summary of six physiological indicators that are often assessed and tracked in emergency departments forms the basis of the NEWS 2 score system. The following parameters are examined: Body temperature, systolic blood pressure, pulse rate, oxygen saturation, respiration rate, and consciousness level[6].

Additionally, the study discovered that clinical sins, statistical modelling, and predictive performance analysis were used to develop specific cutoff values for the mNEWS 2 score. According to a preliminary pilot study, treatment outcomes overlapped in certain score ranges, especially between 0-4 and 5-6 and between patients with scores 7-14 and those with scores of 15 or higher. To ensure that each group had distinct clinical outcomes, it refined risk classification using weighted average risk stratification markers[6].

The computations of accuracy, sensitivity, and specificity revealed a cutoff point at a NEWS 2 score of 15, when prediction performance was maximised provide additional support for these cutoffs. Receiver operating characteristic (ROC) curve analysis was used to validate this, demonstrating that the new classification more accurately reflects mortality risk and intervention requirements.

Another point of strength of this study is the classification of patients into three groups and sub-classification of the third group to three subgroups using the mNEWS 2 score and comparing between them as regards demographic data, biochemical and radiological variables, key clinical differences, treatment given and prognosis helping in accurate risk stratification for more proper time management before more morbidity and mortality.

Limitations

This study is pilot-based and has not been validated externally at multiple locations. Also, it is necessary to add a comparison of other score models (by listing the differences in clinical focus, advantages, and disadvantages of each score model), also include the area under the curve value or ROC graph.

CONCLUSION

The study by Krishnamoorthy et al[6] is a valuable addition to the literature on EPN management. The proposed mNEWS 2 score has the potential to improve risk stratification and treatment outcomes significantly. We look forward to seeing more research in this area, particularly multi-center validation with a larger number of patients and integration with machine learning can further strengthen the accuracy and generalization of mNEWS 2.