Editorial Type: Retrospective Studies
 | 
Online Publication Date: 01 Nov 2018

Utility and Prognostic Significance of Neutrophil-to-Lymphocyte Ratio in Dogs with Septic Peritonitis

DVM,
BVetMed, DACVECC, DECVECC, MRCVS, and
PhD
Article Category: Research Article
Page Range: 351 – 359
DOI: 10.5326/JAAHA-MS-6808
Save
Download PDF

ABSTRACT

Systemic inflammation is known to cause WBC abnormalities, specifically neutrophilia and lymphopenia. The neutrophil-to-lymphocyte ratio (NLR) is a simple and affordable biomarker that has been used in human clinical settings of sepsis but has not been investigated in veterinary species. We evaluated NLR in dogs with septic and nonseptic systemic inflammatory diseases and compared with a healthy dog population. An NLR ≥6 had an 84.39% sensitivity and 86.95% specificity to identify dogs with systemic inflammatory states; however, no ratio distinguished septic and nonseptic causes. The NLR was not associated with length of hospitalization, morbidity based on the acute patient physiologic laboratory evaluation scoring system, or mortality. The disassociation may be due to the retrospective nature of the study, including a restricted population size and acquisition of only a one-time blood sample. NLR is currently of limited use for diagnosis and prognosis in systemic inflammatory states in dogs, and larger, prospective studies are necessary to further evaluate NLR.

Introduction

Sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection and is associated with high morbidity and mortality in both human and veterinary medicine.1–13 Consequently, early and accurate diagnosis of sepsis is of paramount importance for providing timely and adequate interventions.14 Various biomarkers for the diagnosis and prognostication of sepsis have been assessed in veterinary medicine with mixed results.5–8,15–18 Many biomarkers such as C-reactive protein, high mobility group box-1, procalcitonin, and tumor necrosis factor-α require time, may not be readily available, and/or are expensive.7,8,15–17 Fluid cytology and comparison of peritoneal fluid and peripheral blood glucose and lactate concentrations are of diagnostic utility.18 However, little or no fluid may be detected early in the course of septic peritonitis, especially prior to fluid resuscitation in hemodynamically unstable patients or if septic peritoneal effusion is walled off with omentum or mesentery.

Neutrophils are a key component of the innate immune system, resulting in a dramatic increase in the number of circulating neutrophils that migrate to the affected area.19,20 Their role includes the direct killing of pathogens by phagocytosis, release of inflammatory mediators, and activation of T cells.20 However, with continued septic states, neutrophil function is decreased and apoptosis is delayed.21 Prolonged activation of the innate immune system leads to continued inflammation, risk of organ dysfunction, and an impaired adaptive immune system, resulting in the marked loss of both T (CD4+ and CD8+) and B cells via apoptosis.22 Consequently, sepsis may be associated with an immunosuppressive phenotype.

Differentiating sepsis from noninfectious systemic inflammatory response syndrome (NSIRS) based on total leukocyte count is difficult because both diseases induce an inflammatory response.23,24 The neutrophil-to-lymphocyte ratio (NLR) is a readily available, low-cost biomarker that can be calculated from a complete blood count (CBC) to assess the balance of the inflammatory response.25 The NLR is obtained by dividing the total neutrophil count (segmented and band neutrophils) by the number of lymphocytes. Several studies in human medicine have reported that the NLR is an independent predictor of morbidity and mortality in several clinical situations including sepsis, cardiovascular disease, neoplasia, and infectious disease.11,12,26–34

In veterinary medicine, the NLR has largely been evaluated in oncology patients.35–37 To the authors’ knowledge, there is no consensus about the relationship of NLR levels in dogs with sepsis. The NLR could be especially useful in septic patients with cardiovascular compromise (e.g., dehydration, hypovolemia) without detectable peritoneal effusion. The initial objective of this study was to determine the NLR in a population of dogs with septic peritonitis, dogs with NSIRS, and healthy dogs. The secondary aim was to assess if there was any relationship between the NLR and illness severity, length of hospitalization, and outcome as assessed by survival to discharge in the population of dogs with septic peritonitis. We hypothesized that dogs with septic peritonitis would have a higher NLR compared with dogs with NSIRS and clinically healthy dogs. Our secondary hypothesis was that there would be a positive correlation between NLR and illness severity, length of hospitalization, and outcome.

Materials and Methods

Animals

Sepsis group: The medical record database at the University of Illinois Veterinary Teaching Hospital was searched for cases of canine septic peritonitis between July 2012 and March 2017. Dogs were included if they had a diagnosis of septic peritonitis either at presentation or at any time during hospitalization in the intensive care unit. A positive diagnosis was based on the presence of intracellular bacterial or fungal elements on cytological evaluation of abdominal fluid, a positive bacterial or fungal culture, macroscopic visualization of gastrointestinal perforation, or intra-abdominal abscessation with gross peritonitis at time of surgery. A CBC was required to be performed by the hospital laboratory within 24 hr of diagnosis of septic peritonitis.

NSIRS group: The medical record database was also searched for dogs who presented to the Veterinary Teaching Hospital with a surgical gastrointestinal disorder who also satisfied at least 2/4 systemic inflammatory response syndrome (SIRS) criteria during this same time period. The SIRS criteria used for this study were hypothermia (temperature <100.6°F) or hyperthermia (temperature >102.6°F); heart rate >120 beats per minute; respiratory rate >20 breaths per minute; leukopenia (WBC <6 × 109/L), leukocytosis (WBC >16 × 109/L), or >3% band neutrophils.24 Each dog had to undergo surgical exploration to ensure no evidence of sepsis (perforation or abscessation) and have had a CBC performed at the hospital laboratory within 24 hr of diagnosis.

Exclusion criteria for the sepsis and NSIRS groups included dogs receiving chemotherapeutic and/or immunosuppressive medications and dogs diagnosed with hyperadrenocorticism or hypoadrenocorticism. Dogs who had received a transfusion of any allogenic blood product in the 28 days prior to diagnosis of septic peritonitis or NSIRS were also excluded because of concern for immune system modulation.

Control group: A healthy control group consisted of dogs who had presented to the Veterinary Teaching Hospital for routine prescreening blood work. These dogs were deemed healthy based on history and physical examination and were not receiving any medications.

Data Collection

Demographic data including breed, age, gender, reproductive status, and body weight were recorded for all dogs. Blood samples were evaluated using a CBC analyzera. Variables from the CBC obtained from the medical records included hematocrit (HCT), WBC, segmented neutrophil count, band neutrophil count and percentage, lymphocyte count, monocyte count, and platelet concentration. Acute patient physiologic laboratory evaluation (APPLE) scores, in full (APPLEfull) and fast (APPLEfast) form, were retrospectively calculated within 24 hr of diagnosis of septic peritonitis or NSIRS where adequate data was available and expressed as the mortality prediction probability (score 0–1).38 The NLR was calculated for all dogs as NLR = (segmented neutrophils + band neutrophils) / lymphocyte count.

The NLR and APPLE scores were compared in dogs who did and did not proceed for surgical intervention. The cause of septic peritonitis and NSIRS was documented at the time of surgery and recorded. When there were more than five dogs in each disease category, the calculated NLR was compared to assess if there were significant differences in cause for septic peritonitis or NSIRS. Definitive diagnosis for the presence of neoplasia was based on abdominal fluid cytology, fine needle aspirate cytology, or histopathology. The NLR was compared for dogs with and without neoplasia. The length of hospitalization and outcome of the dogs with septic peritonitis and NSIRS undergoing surgery was recorded and assessed for any relationship with NLR. Survival was defined as discharge from the hospital and nonsurvival defined as euthanasia or death during hospitalization.

Statistical Analysis

Dependent variables were analyzed for normality using Kolmogorov-Smirnov and Anderson-Darling normality tests. Initially, outliers were identified from histograms. Statistical identification of outliers is distribution-dependent. Continuous variables (e.g., NLR, breaths, and beats per minute) were analyzed for differences between groups using analysis of variance for data with a Normal (Gaussian) distribution (parametric data) and the Kruskal–Wallis test for continuous data with a distribution that was not transformable to Gaussian (nonparametric data). Post hoc analyses of normally distributed data were carried out with Tukey’s test for comparisons of all possible pairs and Dunnett’s test for comparison of the sepsis and NSIRS groups with the controls. Reference bounds and their 95% confidence bounds for the NLR in healthy dogs were determined for the parametric approach using bootstrap sampling of the NLR value logarithms and exponentiation of the estimated values.39, 40 Logarithms produced a normal distribution and prevented negative estimates of bounds and confidence limits. The receiver operator characteristics area under the curve was used to assess the optimal cutoff NLR to differentiate healthy from unhealthy (NSIRS and septic peritonitis) dogs with their corresponding sensitivity and specificity values. Mann-Whitney U tests were used to test the statistical significance of differences in median values of continuous variables among categories of categorical variables. Spearman’s rank correlation was used to assess association between continuous variables. All statistical analyses were performed using commercial softwareb,c. We distinguished between statistical significance (P < .005) and clinical relevance.

Results

Demographics

Sepsis group: The medical records search identified 115 dogs with the diagnosis of septic peritonitis. Twenty-three were excluded because of a lack of CBC within 24 hr of diagnosis of septic peritonitis, and 12 others were receiving immunosuppressant therapy at time of diagnosis of septic peritonitis. The 80 dogs thus included in the sepsis group comprised 31 breeds. Mixed-breed dogs were overrepresented (n = 18), followed by golden retrievers (n = 7), Labrador retrievers (n = 6), and Yorkshire terriers (n = 4). There were 34 male (42.5%; 30 neutered [88%] and 4 sexually intact [12%]) and 46 female (34 neutered [74%] and 12 sexually intact [26%]) dogs. Dogs in this group had a median age of 7.1 yr (0.42–15 yr). The median weight was 26 kg (2.1–84.1 kg). All dogs satisfied the SIRS criteria.

NSIRS group: NSIRS was identified in 182 dogs. Of these, 84 were excluded because of incomplete or absent blood work within 24 hr of diagnosis of SIRS and surgery, and 5 others who were receiving immunosuppressant therapy at time of presentation. Ninety-three dogs were thus included in the NSIRS group. Of the 33 breeds identified, mixed-breed dogs were overrepresented (n = 17), followed by Labrador retrievers (n = 9), Great Danes (n = 7), German shepherd dogs (n = 5), and golden retrievers (n = 5). There were 61 males (66%, 50 neutered [82%] and 11 sexually intact [18%]) and 32 females (34%, 26 neutered [81%] and 6 sexually intact [19%]) dogs. Dogs in this group had a median age of 5 yr (0.33–14 yr) and a median weight of 28.7 kg (1.7–84 kg).

Control group: One hundred thirty-eight dogs were included in the healthy control group, comprising 37 breeds. Mixed-breed dogs were overrepresented (n = 58), followed by Labrador retrievers (n = 7), golden retrievers (n = 6), Australian shepherds (n = 6), and border collies (n = 6). There were 72 males (52%, 68 neutered [94%] and 4 sexually intact [6%]), 64 females (46%, 53 neutered [83%] and 11 sexually intact [17%]), and 2 of unrecorded sex (2%). The median age was 6 yr (1–15 yr), and the median weight was 21.6 kg (2.2–52.2 kg).

Dogs with septic peritonitis were significantly older than dogs with NSIRS (Table 1). Dogs with NSIRS weighed significantly more than those within the healthy control population, but there was no significant difference between dogs with NSIRS and septic peritonitis. No other significant demographic differences were identified between the three populations (Table 1).

TABLE 1 Population Characteristics and Hematological Parameters for Dogs in the Three Populations
TABLE 1

Clinicopathologic Variables

There were significant differences between the three study populations when assessing HCT, segmented and band neutrophil counts, and lymphocyte counts (Table 1). The APPLEfull scores were calculable for 47 dogs (59%) with septic peritonitis and 43 dogs (46%) with NSIRS. The APPLEfast scores were calculable for 66 dogs (83%) with septic peritonitis and 63 dogs (68%) with NSIRS. Dogs with septic peritonitis had a significantly increased mortality probability based on both APPLEfull and APPLEfast scores compared with dogs with NSIRS (P < .001; Table 1).

NLR: When assessing both mean and median values, there were significant differences for dogs with septic peritonitis or NSIRS when compared with the healthy population (Table 2, Figure 1). However, the mean NLR was not significantly different between dogs with septic peritonitis and NSIRS (P = .009; Table 2).

TABLE 2 Estimated 95% Reference Interval with 95% Confidence Interval by the Bootstrap Method for Neutrophil-to-Lymphocyte Ratio in Healthy Dogs
TABLE 2
FIGURE 1. Box and whisker plot describing the NLR in dogs with septic peritonitis, dogs with NSIRS, and a healthy control population. The boxes represent the 25th–75th percentile, central lines represent the median values, and the whiskers represent the minimum and maximum values. Outliers, values that are 1.5 times removed from the interquartile range, are marked as asterisks. The dashed lines are the confidence interval on the upper reference limit. NLR, neutrophil-to-lymphocyte ratio; NSIRS, noninfectious systemic inflammatory response syndrome.FIGURE 1. Box and whisker plot describing the NLR in dogs with septic peritonitis, dogs with NSIRS, and a healthy control population. The boxes represent the 25th–75th percentile, central lines represent the median values, and the whiskers represent the minimum and maximum values. Outliers, values that are 1.5 times removed from the interquartile range, are marked as asterisks. The dashed lines are the confidence interval on the upper reference limit. NLR, neutrophil-to-lymphocyte ratio; NSIRS, noninfectious systemic inflammatory response syndrome.FIGURE 1. Box and whisker plot describing the NLR in dogs with septic peritonitis, dogs with NSIRS, and a healthy control population. The boxes represent the 25th–75th percentile, central lines represent the median values, and the whiskers represent the minimum and maximum values. Outliers, values that are 1.5 times removed from the interquartile range, are marked as asterisks. The dashed lines are the confidence interval on the upper reference limit. NLR, neutrophil-to-lymphocyte ratio; NSIRS, noninfectious systemic inflammatory response syndrome.
FIGURE 1 Box and whisker plot describing the NLR in dogs with septic peritonitis, dogs with NSIRS, and a healthy control population. The boxes represent the 25th–75th percentile, central lines represent the median values, and the whiskers represent the minimum and maximum values. Outliers, values that are 1.5 times removed from the interquartile range, are marked as asterisks. The dashed lines are the confidence interval on the upper reference limit. NLR, neutrophil-to-lymphocyte ratio; NSIRS, noninfectious systemic inflammatory response syndrome.

Citation: Journal of the American Animal Hospital Association 54, 6; 10.5326/JAAHA-MS-6808

The distributions of NLR for NSIRS and septic peritonitis overlapped 35% of the total range of the NLR distribution for healthy dogs. An optimal NLR cutoff between healthy and unhealthy populations was determined computationally and graphically as NLR = 6, area under the curve = 0.86, sensitivity = 84.39%, and specificity = 86.96% (Figure 2). The NLR could not distinguish dogs with septic peritonitis from those with NSIRS due to ∼100% overlap of the distributions (Figure 1).

FIGURE 2. Receiver operating characteristic curve illustrating the sensitivity and specificity of neutrophil-to-lymphocyte ratio to distinguish dogs with and without systemic inflammation response syndrome. The area under the curve is 0.86 and the point of intersection indicates the optimal cutoff of 6 to identify dogs with systemic inflammation from healthy dogs.FIGURE 2. Receiver operating characteristic curve illustrating the sensitivity and specificity of neutrophil-to-lymphocyte ratio to distinguish dogs with and without systemic inflammation response syndrome. The area under the curve is 0.86 and the point of intersection indicates the optimal cutoff of 6 to identify dogs with systemic inflammation from healthy dogs.FIGURE 2. Receiver operating characteristic curve illustrating the sensitivity and specificity of neutrophil-to-lymphocyte ratio to distinguish dogs with and without systemic inflammation response syndrome. The area under the curve is 0.86 and the point of intersection indicates the optimal cutoff of 6 to identify dogs with systemic inflammation from healthy dogs.
FIGURE 2 Receiver operating characteristic curve illustrating the sensitivity and specificity of neutrophil-to-lymphocyte ratio to distinguish dogs with and without systemic inflammation response syndrome. The area under the curve is 0.86 and the point of intersection indicates the optimal cutoff of 6 to identify dogs with systemic inflammation from healthy dogs.

Citation: Journal of the American Animal Hospital Association 54, 6; 10.5326/JAAHA-MS-6808

The upper bound NLR for normal dogs (13.03) is the 50th percentile of the NLR distribution for all unhealthy dogs. A total of 50/80 (62.5%) dogs with septic peritonitis and 48/93 (51.6%) dogs with NSIRS had a NLR value above the reference bound for normal dogs. NLR values above the reference bound could not distinguish between dogs with septic peritonitis and NSIRS (odds ratio = 0.67 [95% confidence interval: 0.34–1.28, P = .25]).

Thirteen dogs (16%) with septic peritonitis were diagnosed with neoplasia. Only three of these dogs proceeded to surgery. The NLR was not significantly different for dogs with and without neoplasia (P = .054).

Treatment

Fifty-six (70%) dogs from the septic peritonitis group underwent an exploratory celiotomy. The median age of these dogs was 5.83 yr (range 0.5–13 yr), and the median weight was 26.7 kg (range 2.13–84.09 kg). Twenty-four dogs (43%) within this group were not treated surgically. Of these, 20 (83%) dogs were euthanized given the perceived poor prognosis, 1 dog died (4%), and 3 (13%) were discharged against medical advice. The cause of septic peritonitis and NSIRS is summarized in Tables 3 and 4. Dogs who did not proceed to surgery had significantly higher APPLEfull (P < .001) and APPLEfast (P < .001) scores, but the difference in NLR between these populations was not significant (P = .037).

TABLE 3 Etiology of Septic Peritonitis Compared Between Survivors and Nonsurvivors
TABLE 3
TABLE 4 Etiology of Disease in Dogs with Noninfectious Systemic Inflammatory Response Syndrome Confirmed at Time of Surgery
TABLE 4

Outcome and Prognostic Indicators

Forty-two dogs (75%) who underwent surgery for septic peritonitis survived to discharge. Of the 14 nonsurvivors, 4 (29%) experienced cardiorespiratory arrest and 10 (71%) were euthanized. Of the 93 dogs with NSIRS, 89 (96%) survived to discharge following surgery. Of the four nonsurvivors, three (75%) experienced cardiopulmonary arrest and one (25%) was euthanized because of development of a mesenteric and colonic torsion after partial gastrectomy. Following surgery, the 56 dogs diagnosed with septic peritonitis had a median length of hospitalization of 5 (0–17) days. The median length of hospitalization for the dogs in the NSIRS group was 3 (1–8) days. Median lengths of hospitalization between the two groups were statistically different (P < .0001).

When assessing significant differences between variables for survivors and nonsurvivors, only dogs with septic peritonitis were included because of the high survival rate within the NSIRS group. Nonsurvivors had a significantly shorter length of hospitalization, lower segmented neutrophil count, increased HCT, and higher APPLEfast scores compared with survivors. There were no significant differences in age, weight, total WBC, band neutrophil count, lymphocyte count, or NLR between survivors and nonsurvivors in this group (Table 5). The length of hospitalization and NLR in dogs with septic peritonitis were uncorrelated (R = 0.013).

TABLE 5 Population Characteristics, Length of Hospitalization, and Hematological Parameters Between Survivors and Nonsurvivors Among Dogs with Septic Peritonitis Who Underwent Surgical Source Control
TABLE 5

There were weak correlations between the NLR and mortality probability based on APPLEfull score (R = 0.19, P < .001) and APPLEfast score (R = 0.14, P < .001) for the unhealthy dogs.

Discussion

This retrospective study suggests that dogs with systemic inflammation of infectious and noninfectious origins have an increased NLR compared with healthy dogs. Despite dogs with septic peritonitis having a significantly increased total WBC count and band count as compared with dogs with NSIRS, the NLR was not able to differentiate between these two populations.

Zahorec et al. first proposed the use of the NLR as a marker of infection in a clinical setting in human medicine.41 The NLR has since been identified as a simple, readily available biomarker with discriminatory capacity to predict bacteremia and differentiate sepsis from NSIRS in human medicine, which may allow for further understanding of the inflammatory response.23,26,33,41–43 At present, this ratio has been most readily assessed in oncologic patients in the veterinary literature with no previously reported reference intervals (RIs) for normal, healthy dogs.35–37 We were successfully able to determine an RI for a healthy population of dogs to allow for comparison with dogs with systemic inflammation of both infectious and noninfectious etiologies. In contrast to a human study involving healthy patients, our RI was wider with a greater mean NLR value.44

We demonstrated that dogs with septic peritonitis or NSIRS had higher NLR values compared with our healthy control population. The systemic inflammatory response present in both populations is known to result in the development of neutrophilia and lymphopenia. An increased release from the bone marrow and delayed apoptosis of circulating neutrophils typically results in increased numbers of circulating neutrophils of various degrees of maturation.20 However, after initial mobilization and activation, subsequent neutrophils released from the bone marrow have decreased bactericidal function and cytokine production.21 Cortisol mediated apoptosis and margination and redistribution of lymphocytes contribute to the development of lymphopenia.20,21,45 An increased NLR is therefore suggestive of a heightened response to inflammatory insult and raises concern for a relative immunosuppression secondary to continued innate immune system stimulation and down regulation of the adaptive immune system. This is supported by the identification of a positive correlation between NLR, disease severity, and outcome in human medicine.11,12,23,41–43 Furthermore, an elevated NLR has been significantly associated with an increased risk of multiorgan failure and sepsis development in critically ill human patients.33 The significantly higher illness severity scores in the dogs with septic peritonitis compared with those in the NSIRS group may be suggestive of a relatively heightened inflammatory response in these dogs. It was therefore surprising that there were no significant differences in NLR between these two groups in this study.

The absence of a significant difference in NLR between the septic peritonitis and NSIRS groups may be associated with the timing of blood sampling in relation to disease development and progression. A human experimental study identified that the onset of development of lymphopenia following lipopolysaccharide administration occurred before the increase in neutrophil count.19 Therefore, dogs presenting soon after development of septic peritonitis may have a lower NLR compared with those presenting in a more progressed disease state. However, a human clinical study documented an increased NLR despite relative decreases in both neutrophil and lymphocyte counts.43 The retrospective nature of this study, along with dependence of owners to identify clinical signs at home, makes it difficult to accurately assess the effect of duration of sepsis on NLR in dogs.

We were unable to control for antibiotic administration prior to blood collection at the time of diagnosis of septic peritonitis, which could have decreased the NLR in this group. This has been demonstrated in human patients with community-acquired pneumonia.46 Dogs in the septic group were older compared with those in the NSIRS group. Age has been associated with increased neutrophil and decreased lymphocyte counts in dogs.47 This theoretically would have resulted in a more profoundly elevated NLR in the septic peritonitis group, which was not seen. Further studies are required to assess the significance of age on NLR.

The gastrointestinal tract was the most common source of septic peritonitis in this study, which is similar to previous studies.9,10 A gastrointestinal foreign body without perforation was the most common cause of NSIRS. The NLR did not significantly differ in dogs with different underlying disease states, resulting in the development of septic peritonitis or NSIRS. This is consistent with a human study, which did not identify a significant difference in NLR levels between patients with sepsis of pulmonary origin versus abdominal origin.11

Dogs with an underlying neoplastic origin of septic peritonitis did not have a significantly different NLR compared with those without neoplasia. Recent studies in dogs identified an increased NLR to be of prognostic relevance in dogs with soft tissue sarcomas and/or mast cell tumors but not in lymphoma.35–37 The low number of dogs with neoplasia within this study offers an explanation for no significant NLR difference.

The NLR calculated within 24 hr of diagnosis of septic peritonitis was not significantly different in survivors and nonsurvivors following surgery. This is in contrast to certain human studies in which an association was identified with both short- and long-term prognosis.12,42 Conversely, others have identified a lower NLR in patients who died within 5 days of diagnosis of sepsis, which was thought to possibly be related to development of a relative lymphocytosis.27 It is possible that euthanasia may have masked an association between NLR and mortality, especially as these dogs had the highest APPLEfull and APPLEfast scores, suggesting the presence of advanced disease. However, only a poor association was appreciated between NLR and APPLE scores. Serial NLR measurements and a longer follow-up period are required to adequately assess for a significance with outcome in veterinary septic patients.

Total WBC and band neutrophil count were the two components of the leukogram that were able to significantly differentiate dogs with septic peritonitis and NSIRS. The presence of either a leukocytosis or leukopenia is a component of the SIRS criteria. A more profound leukocytosis may be suggestive of sepsis compared with a noninfectious cause.24 However, assessment of total leukocyte count is poorly specific and inconsistent for differentiating sepsis from NSIRS.8,24 Within this study, the mean segmented neutrophil count was higher in the dogs with septic peritonitis and NSIRS compared with the healthy controls; however, the segmented neutrophil count could not differentiate between dogs with sepsis and NSIRS. Dogs with both septic peritonitis and NSIRS had significantly lower mean lymphocyte counts compared with healthy dogs, but again, there was no difference in the mean values between the dogs with septic peritonitis and NSIRS. These findings are consistent with current veterinary literature and suggest assessment of individual WBC is not useful to differentiate sepsis and NSIRS.48

HCT was identified to be significantly lower in dogs with septic peritonitis compared with the NSIRS and healthy control population in this study. This is likely the result of a blunted erythropoietic response and hypoferric state associated with anemia of inflammatory disease along with hemodilution associated with fluid therapy.49 However, no difference was identified in our study between NSIRS and healthy dogs.

Dogs diagnosed with septic peritonitis who did not survive to discharge had significantly higher HCT concentrations compared with survivors. This may be associated with severe extravascular leak or inappropriate IV fluid resuscitation.2 The lower segmented neutrophil count in nonsurvivors is suggestive of acute severe inflammatory disease, likely the consequence of increased margination without sufficient time for bone marrow compensation.50 The difference in APPLEfast scores between the survivors and nonsurvivors supports the idea that mortality in this population was associated with disease severity. However, the role of clinician-based decisions associated with a perceived poor prognosis or financial constraints is not known. We were only able to calculate APPLEfull scores for a small percentage of the sample, and this likely explains the absence of a significant difference between the dogs within the septic peritonitis and NSIRS group.

There are several limitations to the current study, the majority of which are related to its retrospective design. Perhaps the greatest limitation is that the incidence and effect of prior antibiotic administration on the NLR in this study is not known. Second, part of the inclusion criteria was for a CBC to be performed within 24 hr of diagnosis of septic peritonitis or NSIRS. To try to limit the variability in stage of disease progression at presentation, assessment of NLR overtime may be of greater value. It is conceivable that failure of NLR to return to normal limits in the postoperative period is more appropriate for assessment of a relationship with prognosis. Third, our fairly small sample size and population selection may have impacted our results. We included dogs with clinical signs consistent with SIRS criteria and with a surgical abdominal disease process into the NSIRS group so that we could obtain a definitive diagnosis and accurately exclude sepsis. The utility of the NLR with other inflammatory disease states of septic and nonseptic origin and applicability to the larger population of dogs is unknown at present. Finally, we were not able to adequately assess the differences in NLR in dogs with sepsis of varying severity; as per the medical records, only one patient fulfilled the criteria of septic shock. Increased NLRs have been identified in humans with severe sepsis and septic shock compared with sepsis.11 Future studies assessing the differences and utility of NLR in dogs with septic shock compared with sepsis may result in a greater utility of this ratio. Prospective studies to evaluate more comprehensive data sets and to potentially include assessment of inflammatory mediators, red blood cell and platelet indexes, along with NLR, may increase utility of this ratio.

Conclusion

In conclusion, the NLR was able to discriminate dogs with SIRS from a healthy population in this study population. However, the inability to differentiate infectious and noninfectious causes of systemic inflammation, the presence of only a weak correlation with illness severity as determined by APPLE scores, and no significant association with mortality suggest the NLR has limited clinical utility at present.

REFERENCES

  • 1.
    Singer M
    ,
    DeutschmanCS
    ,
    SeymourCW
    , et al    . The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). J Am Med Assoc2016;315(
    8
    ):801–10.
  • 2.
    King LG
    . Postoperative complications and prognostic indicators in dogs and cats with septic peritonitis: 23 cases (1989-1992). J Am Vet Med Assoc1994;204(
    3
    ):407–14.
  • 3.
    Staatz AJ
    ,
    MonnetE
    ,
    SeimHBIII
    . Open peritoneal drainage versus primary closure for the treatment of septic peritonitis in dogs and cats: 42 cases (1993-1999). Vet Surg2002;31(
    2
    ):174–80.
  • 4.
    de Laforcade AM
    ,
    FreemanLM
    ,
    ShawSP
    , et al    . Hemostatic changes in dogs with naturally occurring sepsis. J Vet Intern Med2003;17(
    5
    ):674–9.
  • 5.
    Bentley AM
    ,
    MayhewPD
    ,
    CulpWT
    , et al    . Alterations in the hemostatic profiles of dogs with naturally occurring septic peritonitis. J Vet Emerg Crit Care2013;23(
    1
    ):14–22.
  • 6.
    Cortellini S
    ,
    SethM
    ,
    Kellett-GregoryLM
    . Plasma lactate concentrations in septic peritonitis: A retrospective study of 83 dogs (2007-2012). J Vet Emerg Crit Care2015;25(
    3
    ):388–95.
  • 7.
    DeClue AE
    ,
    SharpCR
    ,
    HarmonM
    . Plasma inflammatory mediator concentrations at ICU admission in dogs with naturally developing sepsis. J Vet Intern Med2012;26(
    3
    ):624–30.
  • 8.
    DeClue AE
    ,
    OsterburK
    ,
    BigioA
    , et al    . Evaluation of serum NT-pCNP as a diagnostic and prognostic biomarker for sepsis in dogs. J Vet Intern Med2011;25(
    3
    ):453–9.
  • 9.
    Bentley AM
    ,
    OttoCM
    ,
    ShoferFS
    . Comparison of dogs with septic peritonitis: 1988–1993 versus 1999–2003. J Vet Emerg Crit Care2007;17(
    4
    ):391–8.
  • 10.
    Culp WT
    ,
    ZeldisTE
    ,
    ReeseMS
    , et al    . Primary bacterial peritonitis in dogs and cats: 24 cases (1990–2006). J Am Vet Med Assoc2009;234(
    7
    ):906–13.
  • 11.
    Liu X
    ,
    ShenY
    ,
    WangH
    , et al    . Prognostic significance of neutrophil-to-lymphocyte ratio in patients with sepsis: a prospective observational study. Mediators Inflamm2016;8191254.
  • 12.
    Hwang SY
    ,
    ShinTG
    ,
    JoIJ
    , et al    . Neutrophil-to-lymphocyte ratio as a prognostic marker in critically-ill septic patients. Am J Emerg Med2017;35(
    2
    ):234–9.
  • 13.
    Boomer JS
    ,
    ToK
    ,
    ChangKC
    , et al    . Immunosuppression in patients who die of sepsis and multiple organ failure. J Am Med Assoc2011;306(
    23
    ):2594–605.
  • 14.
    Vincent JL
    ,
    PereiraAJ
    ,
    GleesonJ
    , et al    . Early management of sepsis. Clin Exp Emerg Med2014;1(
    1
    ):3–7.
  • 15.
    Karlsson I
    ,
    WernerssonS
    ,
    AmbrosenA
    , et al    . Increased concentrations of C-reactive protein but not high-mobility group box 1 in dogs with naturally occurring sepsis. Vet Immunol Immunopathol2013;156(
    1–2
    ):64–72.
  • 16.
    Gebhardt C
    ,
    HirschbergerJ
    ,
    RauS
    , et al    . Use of C-reactive protein to predict outcome in dogs with systemic inflammatory response syndrome or sepsis. J Vet Emerg Crit Care2009;19(
    5
    ):450–8.
  • 17.
    Floras AN
    ,
    HolowaychukMK
    ,
    HodginsDC
    , et al    . Investigation of a commercial ELISA for the detection of canine procalcitonin. J Vet Intern Med2014;28(
    2
    ):599–602.
  • 18.
    Bonczynski JJ
    ,
    LudwigLL
    ,
    BartonLJ
    , et al    . Comparison of peritoneal fluid and peripheral blood pH, bicarbonate, glucose and lactate concentration as a diagnostic tool for septic peritonitis in dogs and cats. Vet Surg2003;32(
    2
    ):161–6.
  • 19.
    Jilma B
    ,
    BlannA
    ,
    PernerstorferT
    , et al    . Regulation of adhesion molecules during human endotoxemia: no acute effects of aspirin. Am J Respir Crit Care Med1999;159(
    3
    ):857–63.
  • 20.
    Lewis DH
    ,
    ChanDL
    ,
    PinheiroD
    , et al    . The immunopathology of sepsis: pathogen recognition, systemic inflammation, the compensatory anti-inflammatory response and regulatory T cells. J Vet Intern Med2012;26(
    3
    ):457–82.
  • 21.
    Hotchkiss RS
    ,
    MonneretG
    ,
    PayenD
    . Sepsis-induced immunosuppression: from cellular dysfunctions to immunotherapy. Nat Rev Immunol2013;13(
    12
    ):862–74.
  • 22.
    Ayala A
    ,
    HerdonCD
    ,
    LehmanDL
    , et al    . Differential induction of apoptosis in lymphoid tissues during sepsis: variation in onset, frequency, and the nature of the mediators. Blood1996;87(
    10
    ):4261–75.
  • 23.
    de Jager CP
    ,
    van WijkPT
    ,
    MathoeraRB
    , et al    . Lymphocytopenia and neutrophil-lymphocyte count ratio predict bacteremia better than conventional infection markers in an emergency care unit. Crit Care2010;14(
    5
    ):R192.
  • 24.
    Hauptman JG
    ,
    WalshawR
    ,
    OlivierNB
    . Evaluation of the sensitivity and specificity of diagnostic criteria for sepsis in dogs. Vet Surg1997;26(
    5
    ):393–7.
  • 25.
    Salciccioli JD
    ,
    MarshallDC
    ,
    PimentelMA
    , et al    . The association between the neutrophil-to-lymphocyte ratio and mortality in critical illness: an observational cohort study. Crit Care2015;19:13.
  • 26.
    Zhang HB
    ,
    ChenJ
    ,
    LanQF
    , et al    . Diagnostic values of red cell distribution width, platelet distribution width and neutrophil-lymphocyte count ratio for sepsis. Exp Ther Med2016;12(
    4
    ):2215–9.
  • 27.
    Riché F
    ,
    GayatE
    ,
    BarthélémyR
    . Reversal of neutrophil-to-lymphocyte count ratio in early versus late death from septic shock. Crit Care2015;19:439.
  • 28.
    Kang MH
    ,
    GoSI
    ,
    SongHN
    , et al    . The prognostic impact of the neutrophil-to-lymphocyte ratio in patients with small-cell lung cancer. Br J Cancer2014;111(
    3
    ):452–60.
  • 29.
    Kalra A
    ,
    WeddJP
    ,
    BambhaKM
    , et al    . Neutrophil-to-lymphocyte ratio correlates with proinflammatory neutrophils and predicts death in low model for end-stage liver disease patients with cirrhosis. Liver Transpl2017;23(
    2
    ):155–65.
  • 30.
    Beliaev AM
    ,
    AngeloN
    ,
    BoothM
    , et al    . Evaluation of neutrophil-to-lymphocyte ratio as a potential biomarker for acute cholecystitis. J Surg Res2017;209:93–101.
  • 31.
    Balta S
    ,
    UsluAU
    ,
    UnluM
    , et al    . The relation between neutrophil–lymphocyte ratio and acute kidney injury. Ren Fail2015;37(
    9
    ):1527–8.
  • 32.
    Ayça B
    ,
    AkınF
    ,
    CelikO
    , et al    . Neutrophil to lymphocyte ratio is related to stent thrombosis and high mortality in patients with acute myocardial infarction. Angiology2015;66(
    6
    ):545–52.
  • 33.
    Akilli NB
    ,
    YortanlıM
    ,
    MutluH
    , et al    . Prognostic importance of neutrophil-lymphocyte ratio in critically ill patients: short-and long-term outcomes. Am J Emerg Med2014;32:1476–80.
  • 34.
    Azab B
    ,
    JaglallN
    ,
    AtallahJP
    , et al    . Neutrophil-lymphocyte ratio as a predictor of adverse outcomes of acute pancreatitis. Pancreatology2011;11(
    4
    ):445–52.
  • 35.
    Macfarlane L
    ,
    MorrisJ
    ,
    PratschkeK
    , et al    . Diagnostic value of neutrophil-lymphocyte and albumin-globulin ratios in canine soft tissue sarcoma. J Small Anim Pract2015;57(
    3
    ):135–41.
  • 36.
    Mutz M
    ,
    BoudreauxB
    ,
    KearneyM
    , et al    . Prognostic value of baseline absolute lymphocyte concentration and neutrophil/lymphocyte ratio in dogs with newly diagnosed multi-centric lymphoma. Vet Comp Oncol2015;13(
    4
    ):337–47.
  • 37.
    Skor O
    ,
    Fuchs-BaumgartingerA
    ,
    TichyA
    , et al    . Pretreatment leukocyte ratios and concentrations as predictors of outcome in dogs with cutaneous mast cell tumours. Vet Comp Onc2016;15(
    4
    ):1333–45.
  • 38.
    Hayes G
    ,
    MathewsK
    ,
    DoigG
    , et al    . The acute patient physiologic and laboratory evaluation (APPLE) score: a severity of illness stratification system for hospitalized dogs. J Vet Intern Med2010;24(
    5
    ):1034–47.
  • 39.
    Solberg HE
    . The IFCC recommendation on estimation of reference intervals. The RefVal Program. Clin Chem Lab Med2004;42(
    7
    ):710–4.
  • 40.
    Dimauro C
    ,
    BonelliP
    ,
    NicolussiP
    , et al    . Estimating clinical chemistry reference values based on an existing data set of unselected animals. Vet J2008;178(
    2
    ):278–81.
  • 41.
    Zahorec R
    . Ratio of neutrophil to lymphocyte counts-rapid and simple parameter of systemic inflammation and stress in critically ill. Bratisl Lek Listy [Sk]2001;102(
    1
    ):5–14.
  • 42.
    Loonen AJ
    ,
    de JagerCP
    ,
    TosseramsJ
    , et al    . Biomarkers and molecular analysis to improve bloodstream infection diagnostics in an emergency care unit. PloS One2014;9(
    1
    ):e87315.
  • 43.
    Gucyetmez B
    ,
    AtalanHK
    . C-reactive protein and hemogram parameters for the non-sepsis systemic inflammatory response syndrome and sepsis: What do they mean?PLoS One2016;11(
    2
    ):e0148699.
  • 44.
    Forget P
    ,
    KhalifaC
    ,
    DefourJP
    , et al    . What is the normal value of the neutrophil-to-lymphocyte ratio?BMC Res Notes2017;10:12.
  • 45.
    Hotchkiss RS
    ,
    SwansonPE
    ,
    FreemanBD
    , et al    . Apoptotic cell death in patients with sepsis, shock, and multiple organ dysfunction. Crit Care Med1999;27(
    7
    ):1230–51.
  • 46.
    de Jager CP
    ,
    WeverPC
    ,
    GemenEF
    , et al    . The neutrophil-lymphocyte count ratio in patients with community-acquired pneumonia. PLoS One2012;7(
    10
    ):e46561.
  • 47.
    Choi SY
    ,
    HwangJS
    ,
    KimIH
    , et al    . Basic data on the hematology, serum biochemistry, urology, and organ weights of beagle dogs. Lab Anim Res2011;27(
    4
    ):283–91.
  • 48.
    Jitpean S
    ,
    PetterssonA
    ,
    HöglundOV
    , et al    . Increased concentrations of serum amyloid A in dogs with sepsis caused by pyometra. BMC Vet Res2014;10:273.
  • 49.
    Nemeth E
    ,
    GanzT
    . Anemia of inflammation. Hematol Oncol Clin North Am2014;28(
    6
    ):671–81.
  • 50.
    Brown MR
    ,
    RogersKS
    . Neutropenia in dogs and cats: a retrospective study of 261 cases. J Am Anim Hosp Assoc2001;37(
    2
    ):131–9.
  1. a

    Cell-Dyn 3700 hematology analyzer; Abbott Diagnostics, Abbott Park, Illinois

  2. b

    SAS 9.4; SAS Foundation, Inc., Cary, North Carolina

  3. c

    SYSTAT 13.1; Systat Software, Inc., San Jose, California

    4. APPLE (acute patient physiologic laboratory evaluation); CBC (complete blood count); HCT (hematocrit); NLR (neutrophil-to-lymphocyte ratio); NSIRS (noninfectious systemic inflammatory response syndrome); RI (reference interval); SIRS (systemic inflammatory response syndrome)
Copyright: © 2018 by American Animal Hospital Association 2018
<bold>FIGURE 1</bold>
FIGURE 1

Box and whisker plot describing the NLR in dogs with septic peritonitis, dogs with NSIRS, and a healthy control population. The boxes represent the 25th–75th percentile, central lines represent the median values, and the whiskers represent the minimum and maximum values. Outliers, values that are 1.5 times removed from the interquartile range, are marked as asterisks. The dashed lines are the confidence interval on the upper reference limit. NLR, neutrophil-to-lymphocyte ratio; NSIRS, noninfectious systemic inflammatory response syndrome.

<bold>FIGURE 2</bold>
FIGURE 2

Receiver operating characteristic curve illustrating the sensitivity and specificity of neutrophil-to-lymphocyte ratio to distinguish dogs with and without systemic inflammation response syndrome. The area under the curve is 0.86 and the point of intersection indicates the optimal cutoff of 6 to identify dogs with systemic inflammation from healthy dogs.

Contributor Notes

Correspondence: efallewellyn@gmail.com (E.A.L.)
*

E. Llewellyn’s present affiliation is Hospital for Small Animals, The Royal (Dick) School of Veterinary Studies, The University of Edinburgh, Edinburgh, United Kingdom.

Accepted: 30 Oct 2017
  • Download PDF