Sepsis in Dogs: Prognostic Indicators and Survival Rates

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-11

Key Takeaways

• Overall 14-day survival in canine SIRS/sepsis is approximately 61%, with septic dogs having a 58% survival rate; however, mortality rises significantly with progression to organ dysfunction and septic shock.
• Protein C deficiency is a potent negative prognostic indicator in septic peritonitis, present in 91% of nonsurvivors versus 13% of survivors, reflecting severe endothelial injury and coagulopathy.
• Serial measurement of C-reactive protein (CRP) is more prognostically valuable than admission values; a decreasing CRP trend over 72 hours predicts survival in 94% of dogs, whereas static or rising CRP is less definitive for predicting death.
• Admission procalcitonin (PCT) concentrations are predictive of organ dysfunction and septic shock, indicating a higher risk of deterioration and warranting intensive monitoring.
• Lactate is a critical prognostic marker, with elevated admission values indicating hypoperfusion; failure of lactate to clear within 12-24 hours of resuscitation signifies persistent hypoperfusion and carries a worse prognosis.
• The number of failing organ systems at admission and their progression over the first 48 hours are strong prognostic indicators, with each additional failing system cumulatively increasing mortality risk.

Sepsis in dogs remains one of the most lethal conditions encountered in small animal emergency and critical care practice. This article reviews the evidence base for prognostic indicators and survival rates in canine sepsis, with emphasis on clinicopathologic variables, biomarker performance, and the physiologic derangements that predict outcome. It is written for practicing veterinarians who need to formulate realistic prognoses, counsel owners, and stratify patients for intensity of monitoring. Treatment protocols are excluded, the focus is on what admission and serial data reveal about likelihood of survival.

The clinical question is direct: which dogs with sepsis are most likely to die, and how accurately can we identify them early? Survival rates reported in the veterinary literature vary with case definition, source population, and study design, but a consistent theme emerges. Mortality in canine sepsis is driven by progression to organ dysfunction, septic shock, and hemostatic failure. Prognostic assessment therefore depends on recognizing these endotypes at presentation and tracking their evolution over the first 48 to 72 hours.

At a Glance

Defining Sepsis and Its Prognostic Framework

The systemic inflammatory response syndrome (SIRS) criteria remain the foundational screening tool for sepsis recognition in dogs, though they lack specificity. Sepsis is defined as SIRS with confirmed or suspected infection. Severe sepsis adds organ dysfunction, and septic shock denotes cardiovascular failure refractory to fluid resuscitation. These distinctions matter prognostically because mortality rises stepwise across the spectrum.

The MSD Veterinary Manual provides the standard clinical reference for these definitions and their application in dogs. Practitioners should apply the criteria consistently at admission and re-evaluate them at fixed intervals, since progression from sepsis to severe sepsis or septic shock is itself a prognostic event.

Survival Rates in Canine Sepsis

Reported survival rates depend heavily on case mix. In a prospective study of 61 dogs with SIRS or sepsis, the 14-day survival rate was 61% overall, with 58% survival in the 48 dogs classified as septic and 69% in the 13 dogs with nonseptic SIRS. These figures, reported by Gebhardt and colleagues in their study of C-reactive protein as a prognostic marker, reflect a mixed population of medical and surgical sepsis sources.

For septic peritonitis specifically, survival is somewhat lower. A prospective cohort of 27 dogs with naturally occurring septic peritonitis reported 59% survival to discharge. The same study, published by Bentley and colleagues as an investigation of hemostatic derangements in septic peritonitis, identified protein C deficiency as a powerful discriminator between survivors and nonsurvivors.

These figures should be interpreted with caution. Single-center studies, variable inclusion criteria, and differences in case management limit direct comparison. The consistent finding across studies is that approximately 40% to 50% of dogs with sepsis do not survive, and that mortality concentrates in patients with organ dysfunction and hemostatic failure.

Biomarkers as Prognostic Indicators

C-Reactive Protein

C-reactive protein (CRP) is an acute phase protein that rises markedly in canine inflammation. The prospective study by Gebhardt and colleagues found no significant relationship between initial serum CRP concentration and survival in dogs with SIRS or sepsis. However, serial measurement told a different story. Changes in CRP concentration over a three-day period correctly predicted survival in 94% of dogs but predicted death in only 30%, with a false positive rate of 22%. The clinical implication is that a falling CRP trajectory supports recovery, while a rising or persistently elevated CRP carries a poor prognosis but with limited sensitivity.

Procalcitonin

Procalcitonin (PCT) has become a standard prognostic biomarker in human sepsis, and evidence in dogs is accumulating. A prospective study of 53 dogs with sepsis, reported by Troia and colleagues in their evaluation of plasma procalcitonin as a predictor of organ dysfunction and outcome, found that baseline PCT concentrations were significantly higher in septic dogs than in healthy controls. Admission PCT was predictive of organ dysfunction and septic shock. Notably, the APPLE fast score calculated at admission was not predictive of sepsis severity, development of multiple organ dysfunction syndrome, or outcome in this cohort.

The practical value of PCT lies in early identification of dogs at risk for deterioration. A dog with elevated PCT at admission warrants more intensive monitoring and earlier intervention, even if clinical signs of organ dysfunction are not yet apparent.

Protein C and Antithrombin

The hemostatic system is intimately involved in sepsis pathogenesis, and its derangements carry prognostic weight. In the septic peritonitis cohort studied by Bentley and colleagues, preoperative protein C deficiency was identified in 91% of nonsurvivors compared with only 13% of survivors. Antithrombin deficiency was equally prevalent in both groups, appearing in 91% of nonsurvivors and 93% of survivors, which limits its value as a standalone prognostic test.

The mechanism is well established. Sepsis triggers widespread endothelial activation, consumption of natural anticoagulants, and a net procoagulant state. Protein C deficiency reflects both consumption and impaired synthesis, and its severity correlates with the extent of endothelial injury. Dogs that survive sepsis typically show recovery of protein C activity over the first days of treatment, whereas persistent deficiency portends a poor outcome.

Physiologic Derangements That Predict Outcome

Hemostatic Failure

Hypercoagulability is a consistent feature of naturally occurring canine sepsis. Thromboelastography in the septic peritonitis cohort demonstrated a hypercoagulable profile in most affected dogs, and this pattern was present regardless of ultimate outcome. The prognostic signal comes not from the presence of hypercoagulability but from the degree of natural anticoagulant depletion that accompanies it.

Organ Dysfunction

The transition from sepsis to severe sepsis is the critical prognostic threshold. Organ dysfunction may involve the kidneys, lungs, liver, or cardiovascular system, and each affected system incrementally reduces survival probability. The RECOVER Initiative guidelines provide evidence-evaluated frameworks for managing the most severe consequence of organ failure, cardiac arrest, and emphasize that survival after cardiopulmonary arrest in septic dogs is poor.

The Role of Serial Assessment

Single time point measurements have limited prognostic power. The CRP data demonstrate that trajectory matters more than baseline value. The same principle likely applies to PCT, protein C, and other biomarkers, though serial data for these analytes in dogs are more limited. A pragmatic approach is to establish a prognostic baseline at admission, repeat key measurements at 24 and 48 hours, and adjust the prognosis based on the direction and magnitude of change.

Prognostic Scoring Systems and Their Limitations

Several scoring systems have been adapted or developed for use in canine sepsis, though none has achieved universal acceptance as a standalone prognostic tool. The APPLE fast score, originally derived for emergency room triage, has been evaluated in septic populations and correlates with illness severity, but its predictive accuracy for individual patient outcomes remains modest. In one study of 53 dogs with sepsis, the APPLE fast score was not predictive of sepsis severity, the development of multiple organ dysfunction syndrome, or outcome, whereas plasma procalcitonin concentrations at admission were predictive of organ dysfunction and septic shock.

The practical limitation of all scoring systems is that they aggregate population-level risk instead of resolving the clinical question of whether a specific patient will survive. A high score justifies aggressive monitoring and treatment escalation, but it does not mandate a particular decision threshold for euthanasia or withdrawal of care. Conversely, a low score does not guarantee survival, particularly when serial assessment reveals deterioration. The clinician should use scoring systems as one input among several, integrating them with trend data, response to resuscitation, and the owner's goals and resources.

Lactate as a Prognostic Marker

Lactate remains the most widely used and best validated prognostic biomarker in emergency and critical care medicine, and this holds for canine sepsis. Admission hyperlactatemia reflects the severity of tissue hypoxia and hypoperfusion, and it correlates with mortality risk across multiple studies. The prognostic value of lactate is strengthened by serial measurement. Failure of lactate to clear within the first 12 to 24 hours of resuscitation identifies patients with persistent hypoperfusion, ongoing anaerobic metabolism, or impaired hepatic clearance, all of which carry a worse prognosis.

The interpretation of lactate must account for its limitations. Hyperlactatemia is not specific to sepsis, and it can arise from hypovolemia, cardiogenic shock, severe anemia, seizures, or neoplasia. In septic patients, elevated lactate may reflect also hypoperfusion but also increased aerobic glycolysis driven by catecholamine surge and inflammatory cytokine signaling. A single elevated lactate value therefore warrants repeated measurement and clinical correlation instead of immediate prognostic pronouncement. The trend, not the absolute value, carries the greater weight in decision-making.

Organ Dysfunction Count and Progression

The number of failing organ systems at admission and the trajectory of organ dysfunction during the first 48 hours are among the strongest prognostic signals in canine sepsis. Renal, hepatic, respiratory, and hemostatic dysfunction each independently increase mortality risk, and the effect is cumulative. A dog with sepsis and single-organ dysfunction has a substantially better outlook than one with three or more failing systems, regardless of the specific organs involved.

The timing of organ dysfunction matters. Early, reversible organ dysfunction that responds to fluid resuscitation and antimicrobial therapy carries a better prognosis than dysfunction that progresses despite treatment. Acute kidney injury that develops after admission, instead of being present at presentation, may reflect either ongoing hypoperfusion or a distinct pathophysiologic process such as acute tubular necrosis from inflammatory injury. Serial monitoring of renal parameters, hepatic enzymes, coagulation status, and respiratory function is therefore essential, also for therapeutic adjustment but for refining the prognostic estimate as the clinical course unfolds.

Hemostatic Failure as a Prognostic Domain

The hemostatic system is both a target and a mirror of the septic inflammatory response. In a prospective study of 27 dogs with naturally occurring septic peritonitis, 16 of 27 dogs survived, and preoperative protein C deficiency was identified in 91% of nonsurvivors compared with 13% of survivors. Antithrombin deficiency was equally prevalent in both groups, but protein C deficiency discriminated outcome with striking clarity. This finding aligns with the broader understanding that protein C depletion reflects endothelial injury, activated coagulation, and impaired endogenous anticoagulant capacity, all of which are central to sepsis-associated organ failure.

The practical implication is that a coagulation panel obtained at admission, including protein C and antithrombin activities, provides prognostic information that complements lactate and organ dysfunction assessment. A dog with septic peritonitis and preoperative protein C deficiency faces a substantially elevated mortality risk, and this should inform the intensity of postoperative monitoring and the discussion with owners about expected outcomes. The absence of protein C deficiency, by contrast, supports a more favorable prognosis even when other parameters are abnormal.

Serial Assessment and Trend-Based Prognosis

Single-point measurements, whether of lactate, procalcitonin, protein C, or C-reactive protein, provide a snapshot that can mislead. The trajectory of these parameters over the first 48 to 72 hours carries more prognostic weight than any admission value. In a study of 61 dogs with systemic inflammatory response syndrome or sepsis, changes in serum C-reactive protein concentrations over a three-day period correctly predicted survival in 94% of dogs, although the test predicted death in only 30% of nonsurvivors, with a false positive rate of 22%. The asymmetry of this result is instructive: a falling C-reactive protein concentration strongly supports recovery, but a rising or static concentration does not reliably predict death.

Serial procalcitonin monitoring may offer similar value. Baseline procalcitonin concentrations predict organ dysfunction and septic shock in dogs with sepsis, and repeated measurement may help identify patients whose inflammatory response is escalating despite treatment. The practical protocol is to obtain baseline values at admission, repeat them at 24 and 48 hours, and interpret the trend in the context of the patient's clinical response, fluid balance, and antimicrobial coverage.

Documenting the Prognostic Assessment

The medical record should capture the prognostic assessment in a structured format that supports clinical decision-making and communication with owners. Record the admission values for lactate, organ dysfunction parameters, and any biomarkers measured, along with the time of sampling. Document the resuscitation response, including changes in lactate, blood pressure, urine output, and mentation over the first 12 to 24 hours. Note the timing and results of serial biomarker measurements, and record the trend explicitly instead of leaving it implicit in the numerical values.

The record should also document the prognostic discussion with the owner, including the estimated survival range based on the available data, the planned reassessment intervals, and the criteria that would trigger escalation or reconsideration of treatment goals. This documentation serves both clinical continuity and medicolegal protection, and it ensures that all members of the care team communicate a consistent prognostic message. Where the evidence base is limited, as it is for many biomarkers in canine sepsis, the record should reflect the uncertainty honestly instead of presenting a false precision.

Recognized Complications and Failure Modes

The dominant failure mode in canine sepsis is progression along the severity spectrum from sepsis without organ dysfunction to severe sepsis and septic shock. Procalcitonin concentrations at admission predict this trajectory, with higher values associated with organ dysfunction and septic shock in dogs plasma procalcitonin concentrations predict organ dysfunction and outcome in dogs with sepsis. Early detection depends on serial physical examination, repeated lactate measurement, and structured organ function assessment instead of a single static evaluation.

Hemostatic failure represents a second major complication. Dogs with septic peritonitis frequently show protein C deficiency, antithrombin deficiency, and hypercoagulability, and preoperative protein C deficiency identifies nonsurvivors with high sensitivity alterations in the hemostatic profiles of dogs with naturally occurring septic peritonitis. Clinicians should screen for petechiae, ecchymoses, prolonged bleeding from venepuncture sites, and unexplained tachycardia out of proportion to perfusion status. Point-of-care coagulation testing and thromboelastography can characterize the phase of coagulopathy, but these tests must be interpreted alongside platelet count and fibrinogen concentration.

Cardiopulmonary arrest is the terminal complication. The RECOVER Initiative veterinary CPR guidelines provide the evidence-evaluated framework for recognizing pre-arrest deterioration and for executing resuscitation when arrest occurs. Anticipation of arrest, preparation of drugs and airway equipment, and clear role assignment before deterioration are more effective than improvisation during the event.

Common Errors and Corrective Actions

A frequent error is anchoring on the admission assessment and failing to re-evaluate. Sepsis is dynamic, and trends matter more than single values. Serial C-reactive protein measurement over three days correctly predicted survival in 94% of dogs but predicted death in only 30%, so a falling CRP supports recovery while a rising or persistently elevated CRP does not reliably confirm a fatal outcome use of C-reactive protein to predict outcome in dogs with systemic inflammatory response syndrome or sepsis. The corrective action is scheduled reassessment at fixed intervals with documentation of trend direction.

A second error is treating lactate or any single biomarker as definitive. No single variable separates survivors from nonsurvivors with acceptable accuracy. The corrective action is multimodal assessment that integrates biomarkers, organ function scores, hemostatic parameters, and perfusion indices.

A third error is under-recognizing the prognostic weight of organ dysfunction count. The number of failing organ systems and the progression from single to multiple organ dysfunction carry more prognostic information than any individual laboratory value. The corrective action is daily systematic organ system review instead of problem-oriented review driven only by abnormal findings.

Limitations of the Current Evidence

The canine sepsis literature is constrained by small cohort sizes, single-center designs, and variable sepsis definitions. The 58% 14-day survival rate reported in one prospective study of 48 septic dogs use of C-reactive protein to predict outcome in dogs with systemic inflammatory response syndrome or sepsis may not generalize across populations with different case mixes, referral patterns, or treatment protocols. Direct extrapolation from human sepsis research is problematic because of species differences in inflammatory mediator biology and in the time course of organ dysfunction.

Expert opinion still differs on the utility of specific biomarkers. Procalcitonin shows promise in dogs, but published data remain limited and assay availability varies plasma procalcitonin concentrations predict organ dysfunction and outcome in dogs with sepsis. Some clinicians favour aggressive early surgical source control in all septic peritonitis cases, while others advocate stabilization before intervention. The evidence base does not resolve this debate.

Escalation and Referral

Referral to a 24-hour emergency and critical care facility is warranted when a dog meets sepsis criteria, when organ dysfunction is present, when hemostatic abnormalities are detected, or when the dog fails to improve within 12 to 24 hours of appropriate treatment. Specialist consultation with a veterinary criticalist should occur early instead of after deterioration. Clinical pathology laboratory involvement is indicated for serial biomarker measurement, coagulation panels, and blood gas analysis.

Regulatory reporting obligations vary by jurisdiction. Veterinarians should consult their local veterinary board and the American Veterinary Medical Association practice resources for guidance applicable to their region. Certain zoonotic or notifiable infections that can present with sepsis may carry reporting requirements under WOAH terrestrial animal health standards, and clinicians should verify current requirements for their species and location.

Troubleshooting Guide

Frequently Asked Questions

How should I interpret a single abnormal biomarker value in an individual septic dog?

A single biomarker measurement rarely determines outcome. In one prospective study, initial C-reactive protein concentrations showed no significant relationship with survival in dogs with SIRS or sepsis, although decreasing concentrations over three days correlated with recovery. Similarly, baseline procalcitonin concentrations predicted organ dysfunction and septic shock, but the authors noted serial monitoring offered additional value. Interpret any single value within the full clinical picture, including perfusion parameters, organ function, and comorbidity burden. A normal biomarker does not override clear clinical deterioration, and an abnormal value does not mandate a hopeless prognosis. Trends across serial measurements carry more weight than isolated readings.

What prognostic information can I provide when advanced biomarker testing is unavailable?

When procalcitonin, protein C, or thromboelastography are not accessible, rely on physical examination, routine clinicopathology, and organ dysfunction assessment. Admission variables such as sepsis score, fibrinogen concentration, and segmented neutrophil count have been associated with survival in critically ill foals, and similar principles apply to dogs. Document perfusion status, mentation, urine output, and serial lactate measurements. The MSD Veterinary Manual provides guidance on interpreting routine laboratory data in critical illness. Prognostic communication should emphasize trend monitoring and response to resuscitation instead of a single test result.

How do I discuss prognosis with an owner when the evidence base is limited?

Acknowledge that published survival rates vary by population and case definition, and that individual outcome cannot be predicted with certainty. Cite the specific survival figures from relevant studies, such as the 58% 14-day survival reported for septic dogs in one cohort, while explaining that these numbers derive from referral populations. Frame the discussion around modifiable factors, including response to fluid therapy, progression of organ dysfunction, and development of hemostatic failure. The AVMA practice resources offer communication frameworks for difficult prognostication conversations. Offer a time-based reassessment plan, for example re-evaluation at 24 and 48 hours, so the owner understands prognosis is dynamic instead of fixed.

How should serial monitoring be structured to refine prognosis during hospitalization?

Reassess perfusion parameters, lactate, organ function scores, and hemostatic markers at defined intervals. In one canine sepsis study, protein C and antithrombin deficiencies were identified preoperatively, with nonsurvivors showing significantly lower protein C activity than survivors. Serial procalcitonin measurements at 24 and 48 hours post-admission provided additional prognostic information beyond baseline values. The RECOVER Initiative guidelines emphasize structured reassessment during critical care. A practical schedule includes admission baseline, reassessment at 12 to 24 hours after resuscitation, and daily evaluation thereafter. Worsening organ dysfunction count or failure of biomarkers to trend toward normal should prompt escalation of monitoring intensity and reconsideration of therapeutic goals.

What should I document in the medical record regarding prognostic assessment?

Record the prognostic indicators used, their numerical values, and the time of measurement. Document the organ dysfunction count, severity classification, and any hemostatic abnormalities identified. Note the basis for the prognostic statement given to the owner, including which studies or guidelines informed your assessment. The AAHA/AAFP fluid therapy guidelines illustrate the standard of documenting monitoring parameters and response to intervention. Include a clear plan for serial reassessment and criteria for escalation or withdrawal of care. If a prognostic score such as APPLE fast was calculated, record the components and the total. This documentation supports continuity across shifts and provides a defensible record of clinical reasoning.

How does the prognostic approach differ when managing septic shock with cardiac arrest risk?

Sepsis complicated by cardiac arrest carries a distinctly grave prognosis, and post-arrest care follows separate consensus guidelines. The RECOVER Initiative provides evidence-evaluated recommendations for resuscitation and post-cardiac arrest management in dogs and cats. Prognostication after return of spontaneous circulation must incorporate neurologic status, duration of arrest, and ongoing hemodynamic instability. Pre-arrest indicators of poor outcome, including refractory hypotension and progressive organ dysfunction, inform discussions about the appropriateness of further resuscitation attempts. Document the arrest event, interventions performed, and the basis for any recommendation to limit further resuscitative efforts. Owner communication should address both survival likelihood and quality of life concerns.

Related Clinical & Scientific Guides

• Toxicology in Emergency Practice: Common Poisons and Diagnostic Approach
• Veterinary Cardiopulmonary Resuscitation: Post-Cardiac Arrest Care
• Fluid Therapy Guidelines for Dogs and Cats: A Practical Update

References and Further Reading

• Use of C-reactive protein to predict outcome in dogs with systemic inflammatory response syndrome or sepsis.. 2009.
• Human genetic and metabolite variation reveals that methylthioadenosine is a prognostic biomarker and an inflammatory regulator in sepsis.. 2017.
• Prognostic value of clinicopathologic variables obtained at admission and effect of antiendotoxin plasma on survival in septic and critically ill foals.. 2006.
• Serum estradiol concentration as a predictor of death in critically ill and injured adults.. 2008.
• Alterations in the hemostatic profiles of dogs with naturally occurring septic peritonitis.. 2013.
• Plasma procalcitonin concentrations predict organ dysfunction and outcome in dogs with sepsis.. 2018.
• RECOVER Initiative Veterinary CPR Guidelines. Veterinary Emergency and Critical Care Society.
• AAHA/AAFP Fluid Therapy Guidelines for Dogs and Cats. AAHA.
• MSD Veterinary Manual, Professional Edition. MSD Veterinary Manual.

Related Articles

• Sepsis in Dogs: Early Recognition and Diagnostic Criteria
• Veterinary Plasma Transfusion: Indications and Administration
• Veterinary Shock: Compensatory Mechanisms and Progression
• Blood Transfusion in Dogs and Cats: Crossmatching and Compatibility
• Electrolyte Emergencies in Dogs and Cats: Recognition and Correction

This article is educational professional reference material for veterinary audiences. It is not a substitute for veterinary diagnosis, individual clinical judgment, current product labeling, or applicable regulatory requirements.