Effectiveness of Mannheim's Peritonitis Index in Patients with Peritonitis Secondary to Hollow Viscus Perforation in a Tertiary Care Hospital

TL;DR — What the literature actually says

1. Background — Why a bedside score matters for HVP peritonitis

Hollow viscus perforation (HVP) — most commonly duodenal, gastric, ileal, appendiceal, or colonic — is the leading cause of secondary peritonitis in Indian emergency surgical admissions. Mortality in India runs 10–30%, considerably higher than the <10% typical in Western centres, driven by delayed presentation, poly-microbial sepsis, and resource constraints on early aggressive resuscitation.

A reliable bedside risk-stratification tool is valuable for four reasons: (1) triage to ICU vs ward, (2) family counselling with an evidence-based prognosis, (3) deciding between damage-control vs definitive surgery, and (4) risk-adjusted institutional audit of outcomes. Existing systems vary in complexity:

• APACHE II / SAPS — ICU-focused, need 24h data collection, not useful in the emergency OT
• SOFA — sepsis-oriented, no intraoperative data
• Boey score — peptic ulcer perforation-specific (3-variable: shock, comorbidity, delay >24h)
• PULP score — specific to perforated peptic ulcers
• POSSUM / p-POSSUM / CR-POSSUM — accurate but arithmetic-heavy; 12 physiological + 6 operative variables + regression equation
• Mannheim Peritonitis Index (MPI) — 8 factors, computable at admission and again intraoperatively

MPI's specific appeal for the Indian tertiary-care setting is that all 8 factors are readily available at bedside or in the OT — no ICU admission required, no ABG needed for the basic score, no waiting on labs beyond admission bloods.

2. The MPI score — what to actually calculate

Eight risk factors. Maximum score 47. Originally derived by Wacha & Linder (1987) on 1,253 German peritonitis patients (1963–1979 cohort); 20 candidate factors were analysed, 8 proved independently prognostic.

Organ failure — the 7-point factor

This is the single highest-weighted factor in the index. A patient with organ failure alone already has MPI = 7, which is in the mild risk tier but at the upper edge. Definitions following Deitch EA (1994):

• Renal: creatinine > 1.6 mg/dL (177 μmol/L), urea > 60 mg/dL (16.7 mmol/L), or oliguria < 20 mL/hr
• Respiratory: PaO₂ < 50 mmHg or PaCO₂ > 50 mmHg
• Cardiovascular: shock (hypodynamic or hyperdynamic)
• Intestinal: bowel paralysis > 24h, or complete mechanical obstruction

3. Risk stratification — what the bands actually mean

The MPI risk bands were defined in the original Wacha & Linder study and replicated across Indian and international validation cohorts. Mortality rates vary across geographies but the gradient is consistent.

4. Predictive performance — 2026 meta-analysis

Pooled performance across 40 studies reviewed in the 2026 PRISMA meta-analysis (Arjun C., Academic Medicine & Surgery, doi:10.62186/001c.162015):

Independent predictors of death in multivariable models include faecal contamination (OR ~26), preoperative organ failure (OR ~51), delayed presentation > 24 h, and malignancy.

5. Novel: MPI + ASA classification combination (2026)

The single most interesting recent development in MPI prognostication — Ma et al., ANZ J Surg, 2026 (PMID 41784287) from the First Affiliated Hospital of Naval Medical University:

• Retrospective, n=248 digestive tract perforation patients (Aug 2021–Aug 2025)
• Overall mortality: 16.5%
• MPI alone: AUC 0.790 (cut-off ≥ 27)
• ASA alone: AUC 0.786
• MPI + ASA combined: AUC 0.904 (sensitivity 87.8%, specificity 81.6%)

The combination outperforms either score alone — a real-world hybrid that addresses one of MPI's main critiques (lack of physiological reserve assessment). ASA adds comorbidity burden; MPI adds peritoneal severity.

6. Practical guidance for clinical use

When to calculate

• At admission — for early triage and family discussion (5 of 8 factors known from history + admission labs)
• Intraoperatively — recalculate after identifying the perforation site, exudate character, and extent. The post-op MPI is the strongest predictor of mortality in most cohorts.
• At 24h — if organ failure develops or resolves, the score changes meaningfully

What to do with the score

• MPI < 21 (mild): Standard post-op ward care; routine antibiotic protocol; consider early oral intake
• MPI 21–29 (moderate): HDU-level monitoring; senior surgical review; consider planned relaparotomy if generalised feculent peritonitis
• MPI > 29 (severe): Mandatory ICU; aggressive resuscitation (fluids, broad-spectrum antibiotics, source control within 6h); damage-control laparotomy if haemodynamically unstable; consider open-abdomen with VAC

Common pitfalls in Indian practice

• Evolution time underestimated — patient may report "yesterday" but actual symptom onset is days earlier. Default to > 24h if unclear.
• Organ failure requires preop labs/ABG — make sure renal function and ABG are drawn at admission; creatinine > 1.6 alone gives 7 points
• Feculent exudate is the biggest single score swing — if intraop finding is fecal peritonitis, that one factor (12 points) often moves the patient up a band on its own. Document it explicitly in the OT note.
• Recalculate postop — if you discover feculent exudate during surgery, the score jumps by 12 points. Communicate the change to the ICU team.

Limitations to acknowledge when teaching

• MPI does not predict morbidity as well as POSSUM — the 2026 meta-analysis is clear on this
• MPI requires intraoperative data for the worst predictor (feculent exudate) — preop-only calculation underestimates risk
• MPI doesn't account for comorbidity burden well — diabetes, immunosuppression, and cirrhosis are scored implicitly via organ failure but not directly (hence the value of MPI+ASA combination)
• The score is not validated for pediatric patients (typically < 15 years excluded from validation cohorts)
• "Organ failure" is binary — same 7 points whether patient needs low-dose vasopressors or is in full anuric multi-organ failure on a ventilator (per Arjun 2026)
• Female sex (5 points) remains controversial — some recent studies suggest this is a historical artefact rather than an independent predictor
• Nechay et al. 2025 (Khirurgiia) specifically caution against using MPI to choose between programmed vs on-demand relaparotomy

7. MPI vs POSSUM — when to use which

Per the 2026 meta-analysis (Chinnappa A.):

"A hybrid approach is suggested: MPI for immediate surgical triage and family counselling, p-POSSUM for retrospective review and institutional benchmarking." — Arjun C., 2026 meta-analysis

Bottom line for a busy Indian surgical unit: use MPI for daily decisions, document POSSUM for monthly audit.

8. The source trail — what to read first

9. How to use this brief

• For teaching rounds: the Section 2 table + Section 3 risk bands print on one side of paper. Use as a bedside reference card.
• For journal club: Sources #1 (Jharkhand 2024) + #9 (2026 meta-analysis) form a 20-minute discussion. They're open-access.
• For departmental protocol drafting: Section 6 "Practical guidance" gives a draft risk-stratified care pathway. Adapt for your hospital's resources.
• For a research project idea: the local calibration question is wide open. Few Indian centres have published their own MPI cut-off. A prospective single-centre cohort of 100–200 patients with mortality endpoint and ROC analysis would be a publishable study. Also: MPI+ASA combination is very new and could be replicated in India.

10. Source audit log

Every source on this page was verified live on 24 June 2026. Method: PubMed search for "Mannheim Peritonitis Index" filtered 2024-2026 (19 results), Scholastica full-text retrieval, Springer/Wiley/PMC direct, Research Square preprint check. No source was accepted from training data alone.

11. Complete bibliography (citation-ready)