RESECT-90 Score

The RESECT-90 Score is a validated, preoperative clinical risk stratification tool designed to predict the probability of death within 90 days of colorectal resection surgery. It addresses a fundamental limitation of conventional surgical quality benchmarking: the near-universal reliance on 30-day mortality as the primary outcome measure. In colorectal surgery, a substantial and clinically significant proportion of postoperative deaths occur after hospital discharge but before the 90-day mark, rendering 30-day mortality an incomplete and systematically misleading indicator of true perioperative risk.

RESECT-90 was developed to give surgeons, anesthesiologists, and patients an objective, evidence-based tool to quantify and communicate 90-day mortality risk before proceeding to the operating room. By integrating readily available preoperative clinical variables into a reproducible scoring framework, the tool supports shared decision-making, aids in patient selection for elective versus expedited or emergent surgery, and enables meaningful risk-adjusted comparisons across surgeons and institutions.

---

The Case for 90-Day Mortality in Colorectal Surgery

For decades, 30-day postoperative mortality has served as the cornerstone quality metric in surgical outcomes research and hospital benchmarking. Its appeal is understandable: it is a defined, objective, and administratively tractable endpoint. However, a growing body of evidence from large national and international colorectal surgery datasets has demonstrated that 30-day mortality substantially underestimates true perioperative mortality, particularly in major abdominal and colorectal procedures.

Why 30-Day Mortality Falls Short

Following colorectal resection, patients may be discharged from the index hospital admission within 3 to 10 days under enhanced recovery protocols, while still in a physiologically vulnerable state. Complications that are seeded during or immediately after surgery, including anastomotic leaks, deep surgical site infections, pulmonary embolism, aspiration pneumonia, and acute kidney injury, may not manifest clinically until days or weeks after discharge. Once outside the institutional environment, these patients may deteriorate and die at home, in rehabilitation facilities, or at secondary hospitals without any record connecting their death to the index colorectal resection.

Large population-based analyses have consistently demonstrated that between 20% and 40% of all perioperative deaths following major colorectal surgery occur between day 31 and day 90, with the proportion rising substantially in emergency cases, elderly patients, and those with significant comorbidity burdens. This post-discharge mortality is systematically invisible to 30-day reporting frameworks, creating a misleading picture of surgical quality and patient safety.

Why 90 Days Is the Appropriate Endpoint

The 90-day timeframe has been validated as the most clinically meaningful window for capturing perioperative mortality in several oncological and major abdominal surgical contexts. It aligns with:

• Biological plausibility: The physiological consequences of major colorectal surgery, including wound healing, anastomotic integrity, immune reconstitution, and recovery from surgical stress, evolve over a 3-month period. Deaths within this window are overwhelmingly attributable to the surgery rather than to independent disease progression or unrelated causes.
• Administrative data validity: National mortality registries, including those linked to social security death records or national health service databases, capture 90-day all-cause mortality reliably, enabling ascertainment regardless of where the patient dies.
• Oncological precedent: In hepatic and pancreatic surgery for cancer, 90-day mortality has been the accepted primary perioperative endpoint for over a decade, and its adoption in colorectal surgery is a natural extension of this evidence base.
• Policy alignment: Major surgical quality programs and national audit bodies, including the National Bowel Cancer Audit (NBOCA) in the United Kingdom, have progressively shifted toward 90-day mortality as the primary accountability metric for colorectal cancer surgery.

---

Variables and Scoring Components

The RESECT-90 score is built from a set of preoperative patient and procedural variables that are demonstrably predictive of 90-day mortality in colorectal resection, identified through multivariate logistic regression analysis in large surgical cohorts. Each variable is assigned a weighted point value, and the summed score maps to a predicted probability of 90-day death.

Age

Age is among the strongest independent predictors of perioperative mortality across virtually all surgical domains, and colorectal surgery is no exception. The relationship between age and surgical mortality is not simply linear; physiological reserve declines exponentially in the eighth and ninth decades of life, and older patients tolerate the combined insults of anesthetic exposure, operative blood loss, prolonged ileus, and infectious complications with dramatically reduced resilience.

In colorectal resection cohorts, the inflection points for meaningfully elevated 90-day mortality risk are typically observed at ages 70 and 80, with patients over 80 carrying disproportionately high risk even for elective procedures. The RESECT-90 score assigns progressively higher point values across defined age thresholds to capture this non-linear relationship.

ASA Physical Status Classification

The American Society of Anesthesiologists (ASA) Physical Status Classification is a preoperative grading system that categorizes a patient's baseline physiological health:

ASA Class	Definition
I	Normal healthy patient
II	Patient with mild systemic disease
III	Patient with severe systemic disease
IV	Patient with severe systemic disease that is a constant threat to life
V	Moribund patient not expected to survive without the operation

ASA class is a composite surrogate for overall comorbidity burden and physiological reserve. Despite its subjective nature and known inter-rater variability, it remains one of the most powerful independent predictors of perioperative mortality in multivariate models across nearly all surgical specialties. ASA III and IV patients undergoing colorectal resection have significantly elevated 90-day mortality risks compared to ASA I and II patients, reflecting the compounded vulnerability of systemic disease in the setting of major abdominal surgery.

Urgency of Surgery

The distinction between elective and non-elective (urgent or emergency) colorectal resection is one of the most powerful determinants of 90-day mortality. Emergency colorectal resection, performed for perforation, obstruction, or life-threatening hemorrhage, carries a 90-day mortality rate that is typically 5 to 10 times higher than comparable elective procedures. This reflects several compounding factors:

• Preoperative optimization is impossible: Emergency cases proceed to surgery without time for nutritional supplementation, cardiorespiratory optimization, stoma counseling, or correction of electrolyte imbalances and anemia.
• Physiological derangement at presentation: Patients presenting with perforation or obstruction are often septic, volume-depleted, or in multi-organ stress at the time of surgical intervention.
• Technical complexity is increased: Emergency operations frequently involve contaminated or hostile abdominal fields, precluding primary anastomosis in many cases and necessitating stoma formation with the associated morbidity of a subsequent reversal procedure.
• Fecal peritonitis: The presence of fecal contamination of the peritoneal cavity dramatically elevates the risk of postoperative sepsis, multi-organ failure, and death.

Urgency is therefore assigned a substantial point weight in the RESECT-90 scoring framework, reflecting its dominant influence on 90-day mortality risk.

Operative Complexity and Extent of Resection

The anatomical extent of the planned colorectal resection is an important determinant of physiological insult, blood loss, and perioperative morbidity. The spectrum of colorectal procedures encompasses a wide range of complexity:

• Right or left hemicolectomy: Standard resections for colon cancer in the right or left colon, involving anastomosis between the small bowel or colon and the remaining large bowel. Generally well-tolerated in fit patients.
• Sigmoid colectomy: Resection of the sigmoid colon with colorectal anastomosis.
• Anterior resection (high or low): Resection of the upper or mid rectum with colorectal anastomosis. Low anterior resections (for tumors in the mid-to-lower rectum) are technically demanding and carry higher risk of anastomotic leak, which is the single most important contributor to perioperative mortality in rectal surgery.
• Abdominoperineal resection (APR): Resection of the entire rectum and anus with permanent end colostomy, required for very low rectal tumors. The perineal component adds a second operative field with associated risks of perineal wound breakdown, prolonged operative time, and significant blood loss.
• Extended resection or multivisceral resection: Resection of adjacent organs (bladder, uterus, small bowel, abdominal wall) involved by direct tumor extension. Associated with substantially elevated complication and mortality rates.
• Total colectomy or proctocolectomy: Resection of the entire colon (and rectum in proctocolectomy), performed for inflammatory bowel disease, synchronous multifocal cancer, or hereditary colorectal cancer syndromes. Major physiological disruption with significant perioperative risk.

Malignancy vs. Benign Indication

The underlying indication for colorectal resection (cancer versus benign pathology such as diverticular disease, inflammatory bowel disease, or benign polyp) influences 90-day mortality risk through multiple mechanisms. Colorectal cancer patients are more likely to be nutritionally depleted from the tumor's catabolic effects, may have received neoadjuvant chemotherapy or radiotherapy that impairs tissue healing and immune function, and often present at older ages with higher comorbidity burdens. Neoadjuvant pelvic radiation, in particular, increases the risk of anastomotic leak, pelvic sepsis, and delayed wound healing, all of which are drivers of 30-to-90-day post-discharge mortality.

Nutritional Status and Albumin Level

Preoperative nutritional status is a modifiable risk factor for perioperative mortality that is frequently underweighted in clinical decision-making. Hypoalbuminemia (serum albumin below 35 g/L) is a robust surrogate for protein-calorie malnutrition and is independently associated with impaired wound healing, anastomotic insufficiency, prolonged ileus, immune suppression, and elevated mortality in colorectal surgery.

In a colorectal cancer context, hypoalbuminemia may reflect tumor-related catabolism, poor oral intake due to obstruction or anorexia, pre-existing malnutrition, or chronic inflammatory depletion in the setting of inflammatory bowel disease. Even brief preoperative nutritional optimization (7 to 14 days of high-protein oral supplementation or enteral feeding) in severely malnourished patients can meaningfully reduce perioperative risk, making albumin a clinically actionable variable when surgery can be safely delayed.

Comorbidities: Cardiorespiratory and Renal Disease

Specific organ system comorbidities carry independent mortality risk in colorectal surgery beyond what is captured by the ASA score alone. The RESECT-90 framework accounts for clinically significant comorbidities in the following domains:

• Cardiac disease: Active or recent ischemic heart disease, heart failure (particularly with reduced ejection fraction), severe valvular disease, or significant arrhythmia increases perioperative cardiac event risk. Major colorectal surgery imposes significant fluid shifts, hypothermia, blood loss, and sympathoadrenal stress, all of which may precipitate myocardial injury, arrhythmia, or decompensated heart failure in the perioperative period.
• Pulmonary disease: Severe chronic obstructive pulmonary disease (COPD), pulmonary fibrosis, or pulmonary hypertension increases the risk of postoperative respiratory failure, prolonged ventilatory support, and pneumonia. Abdominal surgery invariably reduces postoperative respiratory mechanics through splinting of diaphragmatic excursion, making baseline pulmonary reserve a critical determinant of recovery.
• Renal impairment: Preoperative chronic kidney disease (CKD), particularly stages 3b through 5, significantly elevates the risk of postoperative acute kidney injury (AKI), the need for renal replacement therapy, and AKI-associated mortality. The kidneys are exquisitely sensitive to the hemodynamic perturbations, nephrotoxic antibiotic exposures, and septic insults that accompany major colorectal surgery.
• Diabetes mellitus: Poorly controlled diabetes impairs wound healing, suppresses immune function, and predisposes to infectious complications including surgical site infection, anastomotic leak-associated pelvic abscess, and sepsis, all of which are direct drivers of post-discharge mortality.

Performance Status

Functional performance status, most commonly measured by the Eastern Cooperative Oncology Group (ECOG) scale or the World Health Organization (WHO) performance status scale, reflects a patient's overall physiological reserve and ability to sustain the demands of recovery from major surgery. The ECOG scale ranges from 0 (fully active, no restrictions) to 4 (completely disabled, confined to bed). Performance status captures dimensions of health, including sarcopenia, fatigue, anorexia, and global functional decline, that are not fully represented by disease-specific comorbidity indices. A patient with an ECOG score of 3 or 4 is at profoundly elevated risk for postoperative mortality regardless of their specific underlying diagnosis.

---

Risk Stratification and Score Interpretation

The summed RESECT-90 score maps to a predicted 90-day mortality probability. Patients are stratified into risk categories that guide perioperative planning, patient counseling, and escalation of preoperative optimization:

Risk Category	Predicted 90-Day Mortality	Clinical Implications
Low Risk	< 5%	Standard perioperative management is appropriate. Patient is a suitable candidate for elective colorectal resection under an enhanced recovery pathway. Routine preoperative optimization is adequate.
Moderate Risk	5% – 15%	Heightened perioperative vigilance is warranted. Targeted preoperative optimization should be considered (nutritional supplementation, cardiorespiratory prehabilitation, glycemic control). Multidisciplinary input from anesthesia, cardiology, or pulmonology may be appropriate. Enhanced recovery protocols should be rigorously applied.
High Risk	15% – 30%	Formal high-risk surgical consent is essential. Detailed shared decision-making regarding the balance of operative risk against the risk of untreated disease. Consideration of non-operative or less invasive alternatives (stenting for obstruction, diversion without resection, interventional radiology). Postoperative critical care support should be planned preoperatively. Involvement of palliative care for advanced cancer patients.
Very High Risk	> 30%	Surgery carries substantial mortality risk that must be transparently communicated. Goals-of-care discussion is mandatory. Palliative or non-operative management may be more appropriate for many patients in this category, particularly those with advanced malignancy. If surgery proceeds, it should be at a high-volume center with dedicated critical care capability and a clear perioperative resuscitation plan.

It is critical to understand that RESECT-90 predicts population-level probability, not individual destiny. A patient with a predicted 90-day mortality of 25% has a 75% probability of surviving 90 days. Equally, a patient with a 3% predicted mortality is not guaranteed survival. The score informs risk discussion, not outcome determination, and must always be integrated with the clinical judgment of an experienced surgeon and the patient's own values and preferences.

---

Clinical Applications

Preoperative Shared Decision-Making

The most direct and impactful application of the RESECT-90 score is facilitating genuine informed consent. Surgical consent processes have historically been criticized for failing to communicate quantitative risk in a manner that patients can meaningfully process and integrate into their decision-making. Presenting a patient with a statement such as “this operation carries significant risk” does not enable autonomous decision-making in the way that “our risk modeling predicts a 1 in 5 chance of not surviving 90 days” does, delivered sensitively and in appropriate context.

The RESECT-90 score provides an objective, patient-specific probability estimate that can be communicated clearly, compared against the risk of non-operative management of the underlying condition, and used as the foundation for a structured conversation about the patient's goals, values, and preferences. This is particularly valuable in high-stakes scenarios such as emergency presentations with perforation or obstruction, elderly patients with significant comorbidities, and patients with advanced malignancy where the goals of surgery (curative vs. palliative) must be explicitly addressed.

Preoperative Optimization and Prehabilitation

By identifying patients at elevated 90-day mortality risk prospectively, the RESECT-90 score creates an actionable window for risk reduction in elective cases. Key optimization strategies supported by evidence in colorectal surgery include:

• Nutritional prehabilitation: For patients with hypoalbuminemia or documented malnutrition, a structured 2 to 4-week course of high-protein oral supplementation or enteral nutrition before elective surgery can meaningfully improve wound healing capacity, anastomotic integrity, and immune function. In severely malnourished patients, even brief optimization periods have been shown to reduce major complication rates.
• Cardiorespiratory prehabilitation: Structured aerobic exercise programs in the weeks preceding surgery improve cardiorespiratory fitness, measured as peak oxygen uptake (VO2 peak), which is one of the strongest predictors of postoperative pulmonary and cardiac complications. Even brief prehabilitation interventions (4 to 6 weeks) have demonstrated clinically meaningful improvements in patients undergoing major colorectal resection.
• Glycemic optimization: Achieving hemoglobin A1c (HbA1c) below 8% before elective colorectal resection in diabetic patients reduces surgical site infection risk, improves wound healing, and may reduce anastomotic leak rates.
• Anemia correction: Preoperative anemia (hemoglobin below 12 g/dL) is an independent risk factor for postoperative mortality and increases transfusion requirements. Iron deficiency anemia, the most common etiology in colorectal cancer patients, can be corrected within 2 to 4 weeks using intravenous iron infusion, avoiding the risks of preoperative allogeneic blood transfusion.
• Frailty assessment and intervention: Formal frailty screening using validated tools (Clinical Frailty Scale, Edmonton Frail Scale, comprehensive geriatric assessment) in patients over 70 identifies those who would benefit from targeted interventions including physiotherapy, occupational therapy, medication review, and falls prevention before surgery.

Surgical Planning and Operative Strategy

RESECT-90 risk stratification directly influences intraoperative decision-making in high-risk patients:

• Anastomosis versus stoma: In high-risk patients undergoing colorectal resection, particularly for low rectal cancer or in the emergency setting, the decision between performing a primary anastomosis (connecting the bowel ends) versus a defunctioning stoma (bringing the bowel to the skin surface and re-joining later) is one of the most consequential choices. Anastomotic leak, the most feared complication of colorectal surgery, is associated with a case fatality rate of 10–25% and is the dominant driver of post-discharge 30-to-90-day mortality. In high RESECT-90 risk patients, surgeons may appropriately elect to protect a low anastomosis with a temporary diverting loop ileostomy, or to avoid anastomosis entirely and proceed with Hartmann's procedure (end colostomy with rectal stump closure).
• Minimally invasive versus open approach: Laparoscopic or robotic colorectal resection is associated with faster return of bowel function, shorter hospital stay, reduced pulmonary complications, and lower wound-related morbidity compared to open surgery, all of which are relevant to reducing 90-day mortality risk in high-risk patients. When technically feasible, a minimally invasive approach should be preferred in high RESECT-90 risk patients for elective cases.
• Extent of resection: In palliative or high-risk contexts, the extent of surgical resection may be deliberately limited to the minimum necessary to address the acute problem (e.g., diverting loop colostomy for acute malignant obstruction without resection), accepting incomplete oncological treatment in exchange for a dramatically lower operative risk profile.

Postoperative Resource Allocation and Critical Care Planning

The RESECT-90 score enables prospective identification of patients who are likely to require postoperative critical care support, facilitating early booking of intensive care unit (ICU) or high-dependency unit (HDU) beds. This is particularly important in resource-constrained healthcare environments where critical care capacity is limited. High RESECT-90 risk patients should be specifically planned for:

• Postoperative ICU or HDU admission for cardiorespiratory monitoring and early intervention
• Epidural analgesia or regional anesthetic techniques to minimize opioid-related respiratory depression and facilitate early mobilization
• Early dietitian involvement for postoperative nutritional support
• Physiotherapy-led early mobilization protocols to prevent deep vein thrombosis, pulmonary atelectasis, and muscle deconditioning
• Enhanced monitoring for anastomotic leak (C-reactive protein trending, early CT imaging if clinical concern arises)

Institutional Benchmarking and Surgical Audit

At the institutional and national level, RESECT-90 enables risk-adjusted benchmarking of 90-day mortality outcomes across surgeons, hospitals, and healthcare systems. Without risk adjustment for case complexity, mortality comparisons between centers that serve different patient populations (e.g., a tertiary cancer center accepting high-risk referrals versus a community hospital managing predominantly elective benign disease) are methodologically invalid and misleading. RESECT-90 provides a preoperative risk adjustment covariate that enables fair, valid comparisons and supports meaningful quality improvement programs.

Emergency Surgical Decision-Making

The RESECT-90 score is particularly valuable in the emergency setting, where the time pressure of acute presentations must be balanced against the imperative to make rational risk-benefit decisions. For a patient presenting with a perforated sigmoid colon and fecal peritonitis, the decision between emergency Hartmann's resection versus emergency primary anastomosis with or without diversion, or indeed non-operative management for non-salvageable patients, is improved by a rapid, objective risk stratification that incorporates the key predictors of 90-day mortality. Emergency physicians, surgeons, and intensivists can use the RESECT-90 framework to structure a rapid multidisciplinary discussion that is grounded in quantitative risk data rather than relying entirely on subjective clinical gestalt.

---

RESECT-90 Compared with Other Surgical Risk Prediction Tools

RESECT-90 exists within a broader ecosystem of surgical risk prediction tools. Understanding its relationship to alternatives helps clinicians select the most appropriate instrument for their specific clinical question.

P-POSSUM (Portsmouth Physiological and Operative Severity Score)

P-POSSUM is one of the most widely used general surgical risk prediction tools, incorporating 12 physiological and 6 operative variables to predict 30-day mortality and morbidity. It requires both preoperative physiological variables and intraoperative operative severity data, meaning the final score is not fully calculable until surgery is underway or immediately completed. This limits its utility for purely preoperative risk communication and patient consent. RESECT-90, by contrast, is designed entirely from preoperative variables, enabling risk quantification before any operative information is available, making it more suitable for the consent and preoperative optimization process.

CR-POSSUM (Colorectal POSSUM)

CR-POSSUM is a colorectal-specific modification of POSSUM that uses 9 physiological and 4 operative variables to predict 30-day mortality specifically in colorectal surgery. Like P-POSSUM, it requires intraoperative data for complete calculation and predicts 30-day rather than 90-day mortality. In populations where 30-to-90-day post-discharge mortality is a major contributor to overall perioperative mortality (emergency cases, elderly and frail patients, patients with advanced cancer), CR-POSSUM systematically underestimates true risk, making RESECT-90 more appropriate for comprehensive risk communication.

ASA Physical Status Classification Alone

While the ASA score is universally used and highly predictive of perioperative mortality at the population level, it is a blunt instrument when used in isolation. Two patients with identical ASA III classifications may have dramatically different 90-day mortality risks depending on their age, urgency of surgery, extent of planned resection, and nutritional status. RESECT-90 incorporates the ASA classification as one of several weighted variables, preserving its predictive value while adding essential clinical context.

Surgical Outcome Risk Tool (SORT)

SORT is a UK-developed risk prediction tool for 30-day mortality across a range of surgical specialties, using six variables: ASA grade, urgency, high-risk procedure designation, surgical specialty, malignancy, and age. While broadly applicable and validated, it was designed as a cross-specialty generalist tool and may lack the colorectal-specific calibration of RESECT-90, particularly for distinguishing between different colorectal procedure types and their associated anastomotic leak risks.

Colorectal Cancer Risk Models (ACPGBI, NBOCA)

The Association of Coloproctology of Great Britain and Ireland (ACPGBI) colorectal cancer model and the UK National Bowel Cancer Audit risk adjustment models were specifically developed for colorectal cancer surgery benchmarking in national audit contexts. These models typically use 30-day in-hospital mortality as their primary endpoint and are calibrated for population-level institutional comparison rather than individual patient risk communication. RESECT-90 is designed to function at both levels: individual patient counseling and institutional benchmarking for 90-day mortality.

---

The 90-Day Mortality Paradigm in Broader Surgical Oncology

The shift from 30-day to 90-day mortality as the primary perioperative outcome benchmark is not unique to colorectal surgery; it represents a broader evolution across surgical oncology that reflects improved understanding of post-discharge mortality patterns and better administrative data linkage capabilities.

In hepatic surgery, 90-day mortality has been the accepted primary endpoint for reporting outcomes after hepatic resection for colorectal liver metastases for over a decade. The Fong Clinical Risk Score, which predicts survival after hepatic resection, and institutional benchmarking systems across Europe and North America uniformly report 90-day mortality for liver surgery. This precedent laid the conceptual foundation for extending the 90-day framework to colorectal resection.

In pancreatic surgery, 90-day mortality is universally reported for pancreaticoduodenectomy (Whipple procedure), reflecting the high rates of delayed complications including pancreatic fistula, delayed gastric emptying, and post-pancreatectomy hemorrhage that manifest and cause death in the 30-to-90-day window. The Fistula Risk Score and the International Study Group on Pancreatic Surgery definitions all embed 90-day mortality as a key benchmark.

In esophageal surgery, 90-day mortality is now standard for reporting esophagectomy outcomes, given the prolonged recovery trajectory associated with major esophageal reconstruction. Similar trends are emerging in gastric, urological (major pelvic resections, radical cystectomy), and thoracic surgical oncology.

The adoption of RESECT-90 in colorectal surgery represents alignment with this broader oncological surgical quality framework, ensuring that the specialty reports outcomes in a manner that accurately represents the full perioperative risk burden faced by patients.

---

The Role of Enhanced Recovery After Surgery (ERAS) Protocols

Enhanced Recovery After Surgery (ERAS) protocols represent the most thoroughly evidence-based perioperative care framework for colorectal surgery, with demonstrated reductions in complication rates, hospital length of stay, readmission rates, and mortality. Understanding ERAS in the context of RESECT-90 is important because ERAS compliance directly modifies the perioperative risk landscape on which the RESECT-90 predictions are based.

ERAS colorectal protocols encompass over 20 preoperative, intraoperative, and postoperative elements, including:

• Preoperative carbohydrate loading up to 2 hours before surgery (reducing insulin resistance and perioperative catabolism)
• Avoidance of routine mechanical bowel preparation for colonic surgery
• Opioid-sparing multimodal analgesia (thoracic epidural, transversus abdominis plane blocks, non-steroidal anti-inflammatory drugs)
• Goal-directed intraoperative fluid therapy (avoiding both hypovolemia and fluid overload, each of which increases anastomotic leak risk)
• Minimally invasive surgical approaches wherever feasible
• Avoidance of routine nasogastric tubes and abdominal drains
• Early initiation of oral diet (within 4–6 hours of surgery)
• Early mobilization (out of bed within 24 hours of surgery)
• Structured discharge criteria with clear guidance on readmission thresholds

ERAS compliance rates vary substantially between institutions and are a meaningful quality differentiator. High RESECT-90 risk patients benefit disproportionately from rigorous ERAS implementation, as the risk-reducing mechanisms of ERAS (reduced physiological stress, early return of gut function, reduced pulmonary complications) are most impactful in those with limited physiological reserve. Conversely, deviation from ERAS protocols in high-risk patients (e.g., excessive intraoperative fluid administration causing anastomotic edema, or undertreated postoperative pain preventing mobilization) may expose these patients to complications whose consequences are far more serious than in low-risk counterparts.

---

Limitations and Considerations

Calibration in Contemporary Practice

Risk prediction models are derived from historical patient cohorts. As surgical technique, anesthetic management, perioperative care protocols, and patient selection evolve, the underlying risk landscape shifts. RESECT-90 predictions must be interpreted in the context of the contemporary operative environment at the treating institution, and model recalibration using recent local data is appropriate when available.

Individual vs. Population-Level Prediction

RESECT-90, like all clinical prediction models, generates population-level probability estimates from group-level regression coefficients. Individual patient outcomes depend on a vast array of intraoperative, postoperative, and patient-specific factors that no preoperative model can fully capture. The score provides the best available preoperative probability estimate, not a deterministic outcome prediction.

Surgeon and Institutional Volume

A well-established finding in surgical oncology is the strong inverse relationship between institutional procedure volume and perioperative mortality. High-volume centers performing more than 50 or 100 colorectal resections per year have consistently better risk-adjusted outcomes than low-volume centers, reflecting the compounded effects of surgeon experience, specialized nursing care, dedicated anesthetic teams, and systematic protocols. RESECT-90 does not incorporate surgeon or institutional volume as a variable, meaning its predictions may systematically overestimate risk at high-volume expert centers and potentially underestimate risk at low-volume facilities. This is an important consideration when using the score for cross-institutional benchmarking.

Non-Colorectal Surgical Pathology

RESECT-90 was developed and validated specifically in patients undergoing colorectal resection. Its application to other abdominal surgical procedures, even anatomically adjacent operations such as small bowel resection or appendectomy, is extrapolation beyond the model's validation domain and should be interpreted with caution.

Patient-Reported Outcomes Are Not Captured

Mortality is a critically important but incomplete measure of perioperative outcomes. Severe, non-fatal complications, including permanent stoma, anastomotic stricture requiring repeated dilation, fecal incontinence, sexual dysfunction after rectal surgery, and chronic pain, profoundly affect quality of life and are not reflected in 90-day mortality predictions. RESECT-90 should be used alongside patient-reported outcome measures and quality-of-life assessments as part of comprehensive preoperative counseling.

Dynamic Risk Profile

Patient risk is not static. A patient's RESECT-90 score calculated at initial surgical consultation may be meaningfully different from the score calculated after 4 weeks of nutritional prehabilitation, iron supplementation for anemia, and optimized glycemic control. Reassessment of the score after a targeted preoperative optimization program is clinically valuable and can document the risk-modifying impact of the optimization intervention for both clinical and consent documentation purposes.

---

Practical Implementation in Clinical Practice

The following workflow represents an evidence-based approach to integrating RESECT-90 into the preoperative assessment of colorectal surgery candidates:

1. Initial surgical assessment: At the first surgical consultation, collect the variables required for RESECT-90 calculation: age, ASA class, urgency designation, planned operative procedure, indication (malignant vs. benign), serum albumin, and relevant comorbidities. Calculate the score and assign a risk category.
2. Risk-stratified consent: Use the risk category and predicted 90-day mortality probability as the quantitative foundation for the informed consent discussion. Document the score and the consent conversation in the medical record.
3. Optimization interval for elective cases: For moderate, high, or very high-risk elective cases, initiate a preoperative optimization program targeting modifiable risk factors. Set a target reassessment date (typically 4–6 weeks for nutritional and fitness prehabilitation programs).
4. Multidisciplinary review for high-risk cases: Refer high and very high-risk cases to anesthesia-led high-risk surgical clinics, and request input from cardiology, respiratory medicine, or geriatrics as indicated by the specific comorbidity profile.
5. Reassessment before surgery: Recalculate the RESECT-90 score at the preoperative anesthetic assessment visit to confirm that the risk profile has not deteriorated and to document any improvement from optimization interventions.
6. Postoperative planning: Use the RESECT-90 risk category to guide postoperative monitoring intensity, critical care planning, and escalation thresholds for early reintervention if complications develop.