|Year : 2020 | Volume
| Issue : 3 | Page : 457-460
Arterial blood gas as a prognostic indicator in patients with sepsis
Sayan Mukherjee1, Suvrajyoti Das1, Sudipta Mukherjee2, Pralay Shankar Ghosh2, Sanjay Bhattacharya1
1 Department of Microbiology, Tata Medical Center, Kolkata, West Bengal, India
2 Department of Critical Care Medicine, Tata Medical Center, Kolkata, West Bengal, India
|Date of Submission||17-Dec-2019|
|Date of Decision||01-Jul-2020|
|Date of Acceptance||25-Jul-2020|
|Date of Web Publication||4-Nov-2020|
Dr. Sanjay Bhattacharya
Department of Microbiology, Tata Medical Center, 14, Major Arterial Road (EW), Newtown, Kolkata - 700 160, West Bengal
Source of Support: None, Conflict of Interest: None
Abnormal arterial blood gas (ABG) among patients with sepsis is an important prognostic indicator. All-cause mortality was the highest among patients with respiratory acidosis (4/9 = 44.4%), followed by those having metabolic acidosis (3/8 = 37.5%). Median length of hospital and intensive care unit stay was 15.75 days and 6.25 days for those with abnormal ABG and 11 and 3.5 days among those with normal ABG. Median health-care expenditure at the time of discharge or death of the patient was the highest in patients with respiratory acidosis ($14,473) and least in patients with normal ABG ($3,384) (average expenditure among patients with abnormal ABG was [$10,059]).
Keywords: Arterial blood gas, health-care expenditure, hospital stay, mortality, sepsis
|How to cite this article:|
Mukherjee S, Das S, Mukherjee S, Ghosh PS, Bhattacharya S. Arterial blood gas as a prognostic indicator in patients with sepsis. Indian J Med Microbiol 2020;38:457-60
|How to cite this URL:|
Mukherjee S, Das S, Mukherjee S, Ghosh PS, Bhattacharya S. Arterial blood gas as a prognostic indicator in patients with sepsis. Indian J Med Microbiol [serial online] 2020 [cited 2020 Nov 24];38:457-60. Available from: https://www.ijmm.org/text.asp?2020/38/3/457/299806
| ~ Introduction|| |
Patients with sepsis (i.e., infection with organ dysfunction) and cardiorespiratory failure are at the risk of metabolic complications, which could be detected through arterial blood gas (ABG) analysis. The objective of this study was to review: (a) the profile of metabolic abnormalities detected by ABG in patients with clinically suspected sepsis, (b) Correlate ABG profiles with patient outcomes such as mortality, length of the hospital stay and length of the intensive care unit (ICU) stay, (c) correlate ABG with microbiology of positive blood cultures, (d) correlate ABG with organ failures in sepsis and (e) correlate ABG with health-care cost.
| ~ Materials and Methods|| |
The present study was conducted in the ICU, high-dependency unit and surgical ward of an oncology hospital in Eastern India. The study duration was 40 days during the month of May and June 2019. Patients with sepsis were identified as those with suspected infection with at least 2 of the following clinical criteria from a quick sequential organ failure assessment (qSOFA) scoring system: Respiratory rate of 22/min or greater, altered mentation or systolic blood pressure of 100 mmHg or less. Patients with septic shock were clinically identified by a vasopressor requirement to maintain a mean arterial pressure of 65 mm Hg or greater and serum lactate level >2 mmol/L in the absence of hypovolemia. Definition of sepsis (used as inclusion criteria) was the third international consensus definitions for sepsis and septic shock; performance of at least one ABG in the patient and age >18 years. The exclusion criteria were: pediatric patients (≤18 years) and those not fulfilling clinical definition of sepsis as per the international criteria. The criteria of taking blood culture were: (a) Axillary temperature ≥38.4°C (101°F), (b) diagnosis of sepsis and (c) suspected central line associated blood stream infection (CLABSI) or secondary blood stream infection. Blood culture was taken using BactALERT system from Biomerieux, USA. ABG was performed using a GEM Premier 3500 analyzer (Instrumentation Laboratory, Barcelona, Spain).
| ~ Results|| |
During the study period, 230 patients were admitted to the ICU, HDU, surgical ward. Thirty-six patients met the inclusion criteria for sepsis (15.7%) [Figure 1]. The median age of the patients (n = 36) was 58.5 years (range: 19–79 years). The male-to-female ratio of the study cohort was 1:1. Thirteen patients had hematological malignancy and 22 had solid-organ neoplasm, and one had a benign disease. Clinical diagnosis of the study patients included five patients with community-acquired respiratory tract infection; four with hospital acquired respiratory tract infection, two had urosepsis, six had surgical site infection, three patients had CLABSI, and four had non-CLABSI. Blood stream infection included three patients with Escherichia coli bacteremia (sensitive strain), one patient each with Pseudomonas aeruginosa, Serratia marcescens and Burkholderia gladioli bacteraemia (sensitive strain), one patient with Elizabethkingia meningoseptica bacteraemia (resistant isolate), and four cases of Carbapenem-resistant Klebsiella pneumoniae bacteraemia (including one colistin-resistant isolate). One third of the patients (12/36: 33.3%) were either infected or colonised with carbapenem-resistant Enterobacteriaceae/Acinetobacter/Pseudomonas aeuruginosa (K. pneumoniae being the most predominant with 10/36 patients infected or colonised with this pathogen).
|Figure 1: Consort diagram and the profile of the patients with sepsis. Note: Thirty-six patients >18 years of age met the inclusion criteria for sepsis |
Click here to view
The median number of ABG performed in the study cohort (n = 36) was 5.5 (range: 1–29; interquartile range [IQR]: 2.75–10.5). There were 78 instances of organ failures observed in the study cohort. This included 26 (33.3%) instances of circulatory failure, 22 respiratory failure (28.2%), 19 instances of renal failure (24.4%) and 11 hepatic failure (14.1%). The length of hospital stay of the study patients was between 5 and 57 days (IQR: 9–21.5). The median length of ICU stay was 5 days (IQR: 3–7.25). The all-cause 30-day mortality of the total study cohort (n = 36) was 7/36 (19.4%) [Table 1]. The 30-day all-cause mortality in patients not included in the study cohort (n = 194) during the same period was 5/194 (2.6%).
|Table 1: Arterial blood gas profile and clinical outcomes of sepsis patients according to the arterial blood gas performed at the time of intensive care unit admission|
Click here to view
| ~ Discussion|| |
ABG is an important investigation providing information about blood pH (normal value 7.35–7.45), partial pressure of oxygen (PaO2) and carbon dioxide (normal value PaO2: 75–100 mm Hg, PaCO2: 35–45 mm Hg), bicarbonate (HCO3) (normal value: HCO3: 22–26 mEq/L), lactate (normal value: 0.5–2 mEq/L), base excess (normal value −2 to +2 mEq/L), and haemoglobin or haematocrit. In a Swedish study, it was reported that the presence of metabolic alkalosis was independently associated with an increased ICU length of stay. A study from India reported metabolic acidosis to be associated with higher mortality in the ICU. In an Australian study, blood lactate was found to be the strongest predictor for mortality in the multivariate analysis. However, ABG as a test is not readily available in many centres, especially in low- and middle-income countries, either due to resource constraints or inadequacy of appropriately trained and qualified medical personnel to perform and interpret the test. In our setting, the average cost of an ABG was Rs. 895 INR ($13 USD).
The ABG report not only helps in classifying patients based on metabolic abnormalities but also helps in clinical monitoring; decide on specific interventions such as intubation, intravenous fluid infusion, vasopressor administration or hemodialysis. Management of sepsis requires a bundle of interventions which includes taking blood cultures, administration of antibiotics/intravenous fluid/oxygen and testing for blood lactate. Although blood lactate is usually estimated by performing an ABG, ABG per se is not part of the surviving sepsis care bundles. However, the results of ABG are often a major determinant with regard to specific interventions and identifying possible source of the clinical problem. In our study, we found blood lactate to be elevated at presentation to the ICU in 24 patients (24/36 = 67%). Respiratory alkalosis was found to be the most common ABG abnormality, and Gram-negative blood stream infection was the most common cause of sepsis. In our study, respiratory acidosis was found to be the most severe ABG abnormality in terms ICU stay, health-care cost, blood culture positivity and all-cause mortality rate.
ABG is required for: (a) To detect tissue hypoperfusion which is defined as lactate concentration ≥4 mmol/L; (b) To measure blood lactate concentration and targeting resuscitation to normalise lactate; (c) To detect acidosis or academia resulting from tissue hypoperfusion academia for which fluids, vasopressor, HCO3 therapy may be required; (d) To measure hypoxemia and detect severe refractory hypoxemia which may need mechanical ventilation (MV); and (e) Monitor HCO3 level in pH imbalance and HCO3 therapy. Other causes of abnormal ABG were clinically excluded using history, physical examination and other investigations (electrocardiogram and echocardiography for the heart, chest X-ray or computed tomography scan changes for the chest, etc.).
Quick SOFA is a clinical scoring system based on the measurement of blood pressure, respiratory rate and sensorium. ABG analysis gives a more detailed result about metabolic and respiratory parameters of a patient qSOFA. In a study from Australia, SOFA demonstrated significantly greater discrimination for in-hospital mortality (crude area under the receiver operator characteristic [AUROC], 0.753) than systemic inflammatory response syndrome criteria (crude AUROC, 0.589) or qSOFA (crude AUROC, 0.607). In a study from Brazil, the relative risk of death, ICU, and MV need related to qSOFA at admission were 1.83, 0.98 and 1.60, respectively, and its sensitivity was 56.8% for death, 41.4% for ICU need and 53.6% for MV. In another study from Switzerland, the sensitivity of qSOFA was reported to be 31.2% for ICU admission, 30.5% for ICU stay of ≥3 days and 60.0% for mortality at 48 h. Because of the limitations of the clinical scoring systems such as SOFA and qSOFA and the need of metabolic parameter data for clinical management, ABG analysis has become the standard of care in septic patients.
| ~ Conclusion|| |
In conclusion, this small pilot study indicates the value of ABG as a cost effective and rapid (~5 min from sample collection to reporting) diagnostic, monitoring and prognostic tool in the clinical management of patients with sepsis. In resource-constrained settings, it is an important clinical investigation worth considering for installation at all levels of health care which manages patients with sepsis, since it requires significantly less infrastructure (space, manpower, consumables and overheads) and provides rapid results enabling lifesaving supportive care.
This study was conducted as part of the Molecular Medical Microbiology integrated MSc-PhD Program which Tata Medical Center and Indian Institute of Technology, Kharagpur are jointly conducting.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| ~ References|| |
Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al
. The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA 2016;315:801-10.
Kreü S, Jazrawi A, Miller J, Baigi A, Chew M. Alkalosis in critically ill patients with severe sepsis and septic shock. PLoS One 2017;12:e0168563.
Samanta S, Singh RK, Baronia AK, Mishra P, Poddar B, Azim A, et al
. Early pH change predicts intensive care unit mortality. Indian J Crit Care Med 2018;22:697-705.
] [Full text]
Ho KM, Lan NS, Williams TA, Harahsheh Y, Chapman AR, Dobb GJ, et al
. A comparison of prognostic significance of strong ion gap (SIG) with other acid-base markers in the critically ill: A cohort study. J Intensive Care 2016;4:43.
Raith EP, Udy AA, Bailey M, McGloughlin S, MacIsaac C, Bellomo R, et al
. Prognostic accuracy of the SOFA score, SIRS criteria, and qSOFA score for in-hospital mortality among adults with suspected infection admitted to the intensive care unit. JAMA 2017;317:290-300.
Garbero RF, Simões AA, Martins GA, Cruz LVD, von Zuben VGM. SOFA and qSOFA at admission to the emergency department: Diagnostic sensitivity and relation with prognosis in patients with suspected infection. Turk J Emerg Med 2019;19:106-10.
Tusgul S, Carron PN, Yersin B, Calandra T, Dami F. Low sensitivity of qSOFA, SIRS criteria and sepsis definition to identify infected patients at risk of complication in the prehospital setting and at the emergency department triage. Scand J Trauma Resusc Emerg Med 2017;25:108.
Singh V, Khatana S, Gupta P. Blood gas analysis for bedside diagnosis. Natl J Maxillofac Surg 2013;4:136-41.
] [Full text]