A Study on Shock Index in Early Diagnosis of Sepsis in Emergency Department of Tertiary Care Centre of Nepal

Introduction: Sepsis and its consequences, severe sepsis and septic shock is at menace in country like ours where infectious disease are at toll. Early diagnosis and treatment is very important to decrease the morbidity and mortality. Shock index is one of such tool that is very handy in these situations as this is just a mathematical calculation using heart rate(HR) and systolic blood pressure(SBP). The main aim of this study is to find the effectiveness of using SI as an adjunct to blood lactate in diagnosing patients in sepsis. Methods: This was an observational hospital based study conducted at Emergency Department of TUTH, Maharajgunj from 21stjuly 2016 to 13th October 2016, on 104 patients, obtained by purposive sampling method, who had presented to the “Red Area”, aged between 18 to 65 years, who had presented with suspected infection. These patients were screened for severe sepsis u using triage vital signs, basic laboratory tests and an initial serum lactate level. Test characteristics were calculated for hyperlactatemia. I considered the following covariates in my analysis: heart rate >90 beats/min; mean arterial pressure <65 mmHg; respiratory rate > 20 breaths/min; ≥2 SIRS including white blood cell count; SI <0.6; SI 0.6 to 1; SI 1 to 1.4and SI ≥ 1.4. We report sensitivities, specificities, and positive and negative predictive values for the primary outcome. Results: There was apositive correlation between shock index and blood lactate level, r=0.2, n=104, p=0.042.A chi-square test was performed and no relationship was found between SI>=1 and hyperlactemia, X2 (.285, N = 104) = 1, p =.594 and relationship was found between SI >=0.7 and hyperlactemia, X2 (4.1, N = 104) = 1, p =.04. sensitivity and specificity for detecting hyperlactemia of SI>=1, SI.=0.7, SIRS was 84% and 20%, 93% and 0%, 79% and 20% respectively. Negative predictive value of SI>=1, SI.=0.7 and SIRS was 63%, 0% and 57% respectively. Conclusions: There was weak correlation between the lactate level and shock index with statistically significant correlation between the shock index grouped >=0.7 and hyperlactatemia with high sensitivity and very low specificity.


INTRODUCTION
Sepsis and septic shock is a progressive injurious process resulting from systemic inflammatory response to infection. It is a systemic, deleterious host response to infection leading to severe sepsis (acute organ dysfunction secondary to documented or suspected infection) and septic shock (severe sepsis plus hypotension not reversed with fluid resuscitation). Early recognition and prompt resuscitation during the first several hours of severe sepsis and septic shock helps to optimize outcome.
The main aim of this study is to find the effectiveness of using SI as an adjunct to blood lactate in diagnosing patients in sepsis.

METHODS
This was an observational hospital-based study conducted at Emergency Department of TUTH, Maharajgunj from 21 st july 2016 to 13 th October 2016, on 104 patients, who had presented to the "Red Area", aged between 18 to 65 years, obtained by purposive sampling, who had presented with suspected infection. Total patients that had visited the area were 1443 of which 275 were of suspected infection. Depending upon age group, 148 were screened. 104 patients were included in the study excluding incomplete records and referral from other hospital with initial resuscitation.

1.
Age group between 18 and 65

Patients in SIRS
• Two or more of: Temperature >38°C or <36°C, Heart rate >90/min, Respiratory rate >20/min • White blood cell count >12 000/mm 3 or <4000/mm 3  Relation between lactate level and shock index in patients with sepsis.

DISCUSSION
Most screening tools depend on the identification of criteria used to define the systemic inflammatory response syndrome (SIRS). Management tools include serial measurements of blood pressure, heart rate (HR) and lactate levels, tracking organ dysfunction and invasive monitoring. The incidence of sepsis and sepsisrelated mortality is reportedly lower in women, and several hypotheses have been proposed to explain this finding, including the role of sex hormones and sex-related gene polymorphisms associated with immune function. 1 In my study the male population (41.3%) was less than the female population (58.7%) in total cohort and was not comparable to the study by Glickman et al.1 and contradicting the theory. Bernato et al 2 has shown that incidence increases with age which holds true in our study too but in contrary, there is female predominance in the older age group.
The Drug Act of Nepal, implemented in 1978, classifies medicines into several categories and outlines regulation of their use. According to the act, antibiotics may be dispensed by drug sellers only upon receipt of a valid prescription. 3 In this study 25% of the total cohort had not taken any antibiotics. Of 75% of total population had taken antibiotics in one form or the other. There was statistical significance between the population taking antibiotics and not taking antibiotics.This shows that the most of the patients coming to the E.D. would have already consumed some antibiotics in the course of their disease one way or the other.
The most common source of infection was respiratory system (44 out of 104) and the gastrointestinal (19 out of 104) and the genitourinary system (15 out of 104) in order. As compared to the Glickman et al1 where pulmonary system (34%)was the most common one to be affected, and then the genitourinary(14%) and then gastrointestinal system(10%).
The sensitivity of SI >= 1 is 84% and specificity of 20%; the sensitivity of shock index >=0.7 was 93% but specificity of 0%; SIRS criteria had specificity of 79% with specificity of 20%. The incidence of organ dysfunction of >=2 was 64%, 62% and 63% respectively in SI group of >=1 and >=0.6 and SIRS. There was not much of difference in all these three populations.  4 where the negative predictive values for a normal SI and the absence of SIRS criteria for identifying elevated lactate levels were both 0.95, and the sensitivities of SI ≥ 0.7 and of ≥ 2 SIRS criteria were not significantly different. Since many factors affect abnormal vital signs, sensitivity and positive predictive values will vary the true reliability of the findings lies in the negative predictive values of the shock index. In this study shock index less than 1 has more negative predictive value (79%) predictive than shock index less than 0.7 with negative predictive value of 0%. Normal shock index has shown to most cost effective tool to predict and help clinician to prioritize for care. 5 But my study shows that there might be need of categorizing new cut off values for the shock index and further evaluation. The incidence of >=2 organ dysfunction was 64%, 62% and 63% in SI=1, SI.=0.7 and SIRS >=2 respectively. There is not much of difference in the incidence of hyperlactatemia in all these three groups. The negative predictive value of SI>=0.7 was 0% Lactate measurement, important biochemical identifier of sepsis and also predictor of severity, was not considered in the sepsis 3 guidelines for the screening. This guideline has considered SIRS non-specific and now it is no longer used for sepsis recognition/screening.

CONCLUSIONS
There was weak correlation between the lactate level and shock index with statistically significant correlation between the shock index grouped >=0.7 and hyperlactatemia with high sensitivity and very low specificity.