Acute pyelonephritis (APN) spans a broad severity spectrum
from uncomplicated febrile illness to emphysematous pyelonephritis with shock,
creating a persistent need for reliable, bedside risk stratification. This
chapter synthesises evidence on physiologic, disease-specific, and
imaging-based scoring tools. It shows how to assemble them into a practical,
layered pathway that improves triage, timing of imaging, and source control
decisions. We review general sepsis scores (SIRS, SOFA, qSOFA, NEWS2) for early
physiological surveillance; APN-specific clinical models for bacteremia
prediction at first contact; global severity tools (Pitt Bacteremia Score,
Charlson Comorbidity Index) for comorbidity and mortality profiling; and
radiologic frameworks led by the Huang–Tseng CT classification for
emphysematous pyelonephritis. Particular attention is given to a modified NEWS2
(mNEWS2) tailored for EPN, which retains original variables but re-bands risk
and identifies scores ≥15 as a decisive high-risk delimiter linked to ICU need,
nephrectomy, and mortality. We appraise biomarkers that add dynamic signal
pro-calcitonin for bacteremia likelihood, lactate for occult hypoperfusion, and
presepsin as an emerging adjunct and show how trends over 24–72 hours refine
escalation or de-escalation. The chapter translates these elements into an
implementation pathway: screen with NEWS2 or qSOFA; obtain early pro-calcitonin
and lactate; expedite ultrasound or CT when red flags exist (diabetes,
obstruction, acute kidney injury); apply APN specific bacteremia models to
support admission and empiric coverage; and, in confirmed EPN, pair CT class
with mNEWS2 bands to set thresholds for ICU, drainage, and early nephrectomy if
non-responding. Special populations, such as those with diabetes, elderly or
frail patients, transplant recipients, and those with obstruction, are
addressed with lower action thresholds. Finally, we outline emerging
opportunities for precision risk prediction, including machine learning
classifiers to refine early bacteremia detection, radiomics-enhanced CT to
integrate imaging features with clinical data, and transparent validation
standards, moving APN care from static scores to adaptive, multimodal,
precision risk prediction. The result is a clinician-facing roadmap that
accelerates antibiotics, imaging, and decompression, reducing preventable harm
across care systems.
Author(s) Details
Punith Jain R
Department of Urology and Renal Transplantation, Sri Ramachandra Institute
of Higher Education & Research Chennai, India.
Suryaram Aravind
Department of Urology and Renal Transplantation, Sri Ramachandra Institute
of Higher Education & Research Chennai, India.
Vivek Meyyappan
Department of Urology and Renal Transplantation, Sri Ramachandra Institute
of Higher Education & Research Chennai, India.
Velmurugan
Palaniyandi
Department of Urology and Renal Transplantation, Sri Ramachandra Institute
of Higher Education & Research Chennai, India.
Hariharasudhan Sekar
Department of Urology and Renal Transplantation, Sri Ramachandra Institute
of Higher Education & Research Chennai, India.
Sriram Krishnamoorthy
Department of Urology and Renal Transplantation, Sri Ramachandra Institute
of Higher Education & Research Chennai, India.
Please see the book here :- https://doi.org/10.9734/bpi/mono/978-93-47485-93-0/CH5
