Sage Journals: Discover world-class research

Abstract

Introduction:

Shock index is defined as heart rate divided by systolic blood pressure. It is reported as a predictor of morbidity and mortality in trauma and acute myocardial infarction in previous studies. It may be valuable in evaluation of calcium channel blocker poisoning. The objective of this study is to examine the probability of serious outcome based on first measured shock index in patients who presented to emergency department with calcium channel blocker poisoning.

Methods:

A retrospective chart review was conducted on calcium channel blocker poisoning cases in Hong Kong Poison Information Centre from 1 July 2008 to 30 June 2017. Shock index was calculated with blood pressure and pulse measurement at emergency department triage. Odds ratios of various variables for major outcome, mortality, intensive care unit admission, length of stay in acute hospital were calculated by multivariate analysis or negative binomial regression where appropriate. The performance of shock index in predicting major outcome was evaluated with receiver operating characteristic curve.

Results:

A total of 390 cases were identified, of whom 25.1% developed major outcome and 5.6% died. Shock index showed significant association with major outcome (odds ratio: 17.017, 95% confidence interval: 5.521–52.455). The area under the receiver operating characteristic curve for shock index in predicting major outcome was 0.7008 (95% confidence interval: 0.64–0.76).

Conclusion:

Higher shock index is associated with worse patient outcome in calcium channel blocker poisoning. However, shock index alone is not reliable in predicting patient outcome. Further research is needed before shock index can be incorporated for use in patient management in poisoning with calcium channel blocker or other anti-hypertensives.

Keywords

Anti-hypertensive agents emergencies shock toxicology

Introduction

Background

Shock index (SI) is defined as heart rate divided by systolic blood pressure. It is postulated to reflect on stroke volume and cardiac output.¹ Studies on its use in trauma patients identified SI as an independent predictor of increased mortality² and transfusion requirement.^3–5 It is especially useful at an early stage of hypovolaemia with relatively normal blood pressure.^6,7

Use of SI is also investigated in myocardial infarction. Studies show that SI is a predictor of mortality in acute ST-elevation and non-ST-elevation myocardial infarction.^8–11

SI can also help in identifying critically ill patients in the general emergency department setting with heterogeneous patient presentations.¹² Higher SI measured at triage at emergency department is associated with higher rate of hospital admission and inpatient mortality.^1,13,14

However, data on SI in toxicology patients are lacking. Toxicity from cardiovascular drugs causes significant mortality. Cardiovascular drug ranked as the second commonest agent involved in poisoning deaths reported to poison centres in the United States in 2015.¹⁵ Among cardiovascular drugs, calcium channel blockers accounted for 40% on the death cases.¹⁵ Hence calcium channel blocker poisoning is chosen for investigation in this study.

Calcium channel blockers inhibit voltage-gated L-type calcium channels and decrease calcium influx into myocardial cells and vascular smooth muscle cells. Hypotension is a feature of significant toxicity and is due to the exaggeration of the therapeutic effects, namely a combination of negative myocardial inotropy, chronotropy and vasodilatation.¹⁶

We postulate that in patients with calcium channel blocker poisoning, higher values of SI reflect poorer haemodynamic status and hence can predict poorer outcomes.

Study objective

The objective of this study is to examine the probability of serious outcome based on first-measured SI in patients who presented to emergency department with calcium channel blocker poisoning.

Methods

Study design and setting

This is a retrospective chart review of cases in clinical database of Hong Kong Poison Information Centre (HKPIC), from 1 July 2008 to 30 June 2017.

HKPIC provides consultation on poison information and management advice to all healthcare professionals in Hong Kong. It also receives poison reporting from Hospital Authority. It receives more than 3000 consultations and reporting cases each year.

The database (Poison Information and Clinical Manage-ment System – PICMS) contains consultation and reporting cases. The data of each poisoning case were fed to PICMS according to information obtained from the consulting health care professional, medial record, electronic patient record (ePR) of Hospital Authority and other relevant sources.

Description of study subjects

Patients were recruited through data search in the database of HKPIC. Poisoning cases were collected to the database through phone consultation or Hospital Authority poison reporting system.

Inclusion criteria

All patients aged 18 years or older and who were labelled as calcium channel blocker poisoning during the period were included in the review. They were identified by data search in PICMS database.

Exclusion criteria

Patients with unmeasurable blood pressure at triage, incomplete record of blood pressure and heart rate at triage, incomplete clinical records and those with subsequent confirmed non-exposure are excluded. Prisoners were not included in this study.

Variable and outcome

Variables that were recorded for each patient encounter include demographic data, toxin data, and clinical data. Demographic data include age and sex of patient. Toxin data include name and dose of calcium channel blocker ingestion, other co-ingested substance, and the reason of exposure. Clinical data include the first set of blood pressure and pulse measured at triage of emergency department, treatment received, disposition and outcome of the patient.

Patient outcome was classified by clinical toxicologists in HKPIC with reference to the American Association of Poison Control Centres’ National Poison Data System. It is divided into five categories: no effect, mild effect, moderate effect, major effect or death, as listed in Table 1.¹⁷ Patient outcome was fed to the PICMS database for every case according to the available clinical information.

Table 1.

Definition of clinical outcome.

No effect	The patient did not develop any signs or symptoms.
Mild effect	The patient developed some signs or symptoms that were minimally bothersome and generally resolved rapidly with no residual disability or disfigurement.
Moderate effect	The patient exhibited signs or symptoms that were more pronounced, more prolonged or more systemic in nature than mild effect. Usually, some form of treatment is indicated. Symptoms were not life-threatening and the patient had no residual disability or disfigurement.
Major effect	The patient exhibited signs or symptoms that were life-threatening or resulted in significant residual disability or disfigurement.
Death	The patient died.

Primary and secondary outcomes

Primary outcome of this study is the occurrence of major effect or death (major outcome).

Secondary outcomes include mortality, admission to intensive care unit (ICU) and length of stay in acute hospital ward.

Sample size and calculation

A previous study demonstrated area under receiver operating characteristic (ROC) curve of SI in predicting major injury in trauma patients to be 0.6.¹⁸ Prevalence of major effect or death in calcium channel blocker poisoning is estimated to be 0.25 based on previous experience. Sample size calculation is based on a formula proposed by Obuchowski and Mcclish.¹⁹ Assuming margin 0.1, alpha 0.05, beta 0.2, required sample size for comparison of area under curve is at least 332.

Data collection

The clinical records including Accident and Emergency Department attendance record and in-patient discharge summaries were retrieved from computer database PICMS. The blood pressure and heart rate at triage were recorded on paper record of each patient in emergency department, which was scanned and stored into the database. The blood pressure and heart rate for each patient were manually extracted from scanned paper records to a spreadsheet.

Only data relevant to study purpose were recorded, which include basic patient demographic information including age, sex and other relevant clinical information.

Statistical analysis

Univariate and multivariate regression analysis was performed on association of variables with the primary outcome, mortality and ICU admission. Length of stay in acute hospital was analysed using negative binomial regression, and mortality cases were assigned to have the longest length of stay. P value less than 0.05 is defined as statistically significant.

The optimal cutoff for SI in predicting serious outcome in calcium channel blocker poisoning is not known, so the performance of SI in predicting serious outcome in calcium channel blocker poisoning was evaluated by receiver operating characteristic curve.

The statistical analysis was performed using SPSS, version 20.0.

Results

During the period of 1 July 2008 to 30 June 2017, 411 cases of calcium channel blocker poisoning were recorded in the database. Twenty-one of those cases were excluded: 14 had outcome unrelated to calcium channel blocker exposure, 1 had undetectable blood pressure at emergency department, 6 cases were not presented to emergency departments; so 390 cases were included in the final analysis. Patients’ characteristics are shown in Table 2.

Table 2.

Patients’ characteristics.

Variable	Total (N = 390)
Age*	64 (18)
Sex, female	218 (55.9%)
Suicidal intent	222 (56.9%)
Co-ingestion of anti-hypertensives	138 (35.4%)
Pre-existing hypertension	301 (77.2%)
SI at triage*	0.757 (0.269)
Major outcome	98 (25.1%)
Mortality	22 (5.6%)
ICU admission	129 (33.1%)
Length of stay in acute hospital (days)^†	3 (1–5)

SI: shock index; ICU: intensive care unit.

Mean (SD).

†

Median (Q1–Q3).

Univariate analysis on association of variables with major outcome, mortality and ICU admission was performed. Age, pre-existing hypertension, anti-hypertensives co-ingestion, suicidal intent and SI were considered to have potential association with the outcomes and analysed. It showed that suicidal intent and SI had significant association with all three outcome measures (Table 3).

Table 3.

Univariate analysis of variables related to major outcome, mortality and ICU admission.

Variable	Major outcome		Mortality		ICU admission
Variable	OR (95% CI)	P value	OR (95% CI)	P value	OR (95% CI)	P value
Age	0.987 (0.974–0.999)	0.04	1.018 (0.993–1.044)	0.166	0.975 (0.963–0.987)	<0.001
Pre-existing hypertension	1.029 (0.595–1.778)	0.919	1.931 (0.558–6.683)	0.299	1.26 (0.753–2.11)	0.378
Co-ingestion of anti-hypertensives	1.289 (0.804–2.065)	0.292	1.046 (0.428–2.559)	0.921	1.126 (0.726–1.746)	0.596
Suicidal intent	9.198 (4.716–17.940)	<0.001	5.148 (1.497–17.698)	0.009	13.048 (6.99–24.357)	<0.001
SI	20.689 (7.495–57.111)	<0.001	5.419 (1.571–18.691)	0.007	21.74 (8.131–58.131)	<0.001

ICU: intensive care unit; OR: odds ratio; CI: confidence interval; SI: shock index.

In the multivariate analysis, final model was created for major outcome, mortality and ICU admission by the backward conditional method. The results are shown in Table 4.

Table 4.

Multivariate analysis of variables.

Variable	Major outcome		Mortality		ICU admission
Variable	OR (95% CI)	P value	OR (95% CI)	P value	OR (95% CI)	P value
Age	1.015 (0.999–1.031)	0.063	1.049 (1.017–1.082)	0.002
Co-ingestion of anti-hypertensives	1.968 (1.128–3.434)	0.017			1.756 (1.024–3.009)	0.041
Suicidal intent	10.521 (4.914–22.526)	<0.001	8.904 (2.306–34.377)	0.002	12.070 (6.233–23.374)	<0.001
SI	17.017 (5.521–52.455)	<0.001	8.794 (2.391–32.342)	0.001	12.596 (4.343–36.531)	<0.001

ICU: intensive care unit; OR: odds ratio; CI: confidence interval; SI: shock index.

SI had the strongest association with major outcome with an odds ratio of 17.017 (95% confidence interval (CI): 5.521–52.455). Suicidal intent had odds ratio of 10.521 (95% CI: 4.914–22.526). Anti-hypertensives co-ingestion had odds ratio of 1.968 (95% CI: 1.128–3.434). Age was only borderline significant with odds ratio 1.015 (95% CI: 0.999–1.031).

Length of stay in acute hospital was analysed using negative binomial regression, and mortality cases were assigned to have the longest length of stay. SI had the strongest association with length of stay with odds ratio 4.177 (95% CI: 2.3337–7.4779; Table 5).

Table 5.

Length of stay in acute hospital (negative binomial regression).

Variable	OR (95% CI)	P value
Age	1.022 (1.0124–1.0317)	<0.001
Suicidal intent	3.1231 (2.2087–4.416)	<0.001
SI	4.1775 (2.3337–7.4779)	<0.001

OR: odds ratio; CI: confidence interval; SI: shock index.

Figure 1 shows the receiver operating characteristic curve for SI in predicting major outcome. SI had area under the curve of 0.7008 (95% CI: 0.64–0.76).

Figure 1.

Receiver operating characteristic curve of major outcome prediction with shock index.

The operating characteristic of SI is included in Table 6.

Table 6.

Operating characteristics of SI in predicting major outcome.

Shock index	Sensitivity	Specificity	Positive predictive value	Negative predictive value
0.5	0.949	0.178	0.279	0.912
0.8	0.582	0.685	0.383	0.83
1.0	0.316	0.894	0.5	0.796
1.2	0.133	0.976	0.65	0.77
1.5	0.041	1	1	0.756

SI: shock index.

Discussion

Calcium channel blocker poisoning causes significant morbidity and mortality. Prompt recognition of patients who are more likely to develop serious outcome is necessary, as it can guide whether to give more aggressive gastrointestinal decontamination treatment and intensive monitoring.

In this study, SI is calculated by systolic blood pressure and heart rate measured at triage in patients with calcium channel blocker poisoning presenting to emergency department. The association of SI and major outcome is examined with multivariate logistic regression. SI was demonstrated to have significant association with major outcome, mortality, ICU admission, and length of hospital stay. This is compatible with our proposed theory that greater value of SI reflects worse haemodynamic status of the patient and hence predicts poorer outcome in patients.

In the analysis of the operating characteristic, SI had only fair performance as a sole predictor of major outcome. This suggests that although SI is associated with serious outcome, there are other important determinants. One possible factor is that a single SI measurement alone does not have sufficient predictive power to predict patient outcome. In a previous study conducted in trauma patients, the change of SI from pre-hospital to emergency department is also a sensitive predictor of mortality.²⁰

Anti-hypertensives co-ingestion was only shown to have mild association with outcome. There are different anti-hypertensives each with different toxicity profiles, and they affect the blood pressure and heart rate to different extents. Because of small sample size, only the presence or absence of anti-hypertensives co-ingestion was counted in the analysis. The significance of anti-hypertensive co-ingestion in predicting outcome may be underestimated.

This study has several limitations.

First, the blood pressure and pulse were usually measured by nurses at triage using an automated machine. In extremes of blood pressure or irregular pulse, this could lead to inaccurate measurements. This retrospective study cannot prevent this source of error.

Second, the poisoning cases in the database were collected from the treating physicians on a voluntary basis through reporting and consultation. Less severe cases may not be collected.

Third, calcium channel blockers can be classified into dihydropyridines and non-dihydropyridines. They produce different clinical features in overdose. Dihydropyridines produce reflex tachycardia while non-dihydropyridines such as diltiazem and verapamil produce bradycardia.^16,21 The issue is further complicated as there are cases with ingestion of both dihydropyridines and non-dihydropyridines. Subgroup analysis could not be done due to the small size of the group with non-dihydropyridines ingestion.

Fourth, due to the retrospective nature of this study, the data on dosage of calcium channel blockers ingestion is incomplete. In many cases, the dosage ingested was not reported or was uncertain. While it is clear that the toxicity of calcium channel blockers is dose-dependent, the relationship between dosage of calcium channel blockers ingested and SI and outcome cannot be shown in this study because of this limitation.

In conclusion, higher SI is associated with more serious patient outcome in calcium channel blocker poisoning. However, SI alone is not reliable in predicting patient outcome. Further research is needed before SI can be incorporated for use in patient management in calcium channel blocker poisoning or other anti-hypertensives poisoning.

Footnotes

Author contributions

M.T.L. researched the literature, conceived the study, gained ethical approval, collected data and wrote the first draft of the manuscript. Both authors were involved in development of the research protocol and data analysis. Both authors reviewed and edited the manuscript and approved the final manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

Availability of data

The data sets generated and/or analysed during the current study are available from the corresponding author.

Informed consent

Informed consent was waived by the Hospital Authority Research Ethics Committee.

Ethical approval

The study was approved by the Hospital Authority Research Ethics Committee, Kowloon Central/Kowloon East. (KC/KE-17-0178/ER-3), and the study protocol comply with the Declaration of Helsinki.

ORCID iDs

Man Ting Lau

Cheung Lun William Wong

References

Balhara

Hsieh

Hamade

, et al. Clinical metrics in emergency medicine: the shock index and the probability of hospital admission and inpatient mortality. Emerg Med J 2017; 34(2): 89–94.

McNab

Burns

Bhullar

, et al. A prehospital shock index for trauma correlates with measures of hospital resource use and mortality. Surgery 2012; 152(3): 473–476.

Olaussen

Blackburn

Mitra

, et al. Review article: shock index for prediction of critical bleeding post-trauma – a systematic review. Emerg Med Australas 2014; 26(3): 223–228.

Mutschler

Nienaber

Münzberg

, et al. The shock index revisited: a fast guide to transfusion requirement? A retrospective analysis on 21,853 patients derived from the TraumaRegister DGU. Crit Care 2013; 17(4): R172.

Mitra

Fitzgerald

Chan

The utility of a shock index ⩾1 as an indication for pre-hospital oxygen carrier administration in major trauma. Injury 2013; 45: 61–65.

Vandromme

Griffin

Kerby

, et al. Identifying risk for massive transfusion in the relatively normotensive patient: utility of the prehospital shock index. J Trauma 2011; 70(2): 384–388; discussion 388–390.

Birkhahn

Gaeta

Terry

, et al. Shock index in diagnosing early acute hypovolemia. Am J Emerg Med 2005; 23(3): 323–326.

Kobayashi

Misumida

Luger

, et al. Shock index as a predictor for in-hospital mortality in patients with non-ST-segment elevation myocardial infarction. Cardiovasc Revasc Med 2016; 17(4): 225–228.

Bilkova

Motovska

Widimsky

, et al. Shock index: a simple clinical parameter for quick mortality risk assessment in acute myocardial infarction. Can J Cardiol 2011; 27(6): 739–742.

10.

Huang

Yang

Zhu

, et al. Usefulness of the admission shock index for predicting short-term outcomes in patients with ST-segment elevation myocardial infarction. Am J Cardiol 2014; 114(9): 1315–1321.

11.

Gouda

Saad

Al-Daydamony

MM.

Modified shock index as a predictor of in-hospital outcome in cases of ST-segment elevation myocardial infarction treated with primary percutaneous coronary intervention. J Cardiol Curr Res 2016; 7(4): 00255.

12.

Rady

Smithline

Blake

, et al. A comparison of the shock index and conventional vital signs to identify acute, critical illness in the emergency department. Ann Emerg Med. 1994; 24(4): 685–690.

13.

Torabi

Mirafzal

Rastegari

, et al. Association of triage time shock index, modified shock index, and age shock index with mortality in emergency severity index level 2 patients. Am J Emerg Med 2016; 34(1): 63–68.

14.

Torabi

Moeinaddini

Mirafzal

, et al. Shock index, modified shock index, and age shock index for prediction of mortality in emergency severity index level 3. Am J Emerg Med 2016; 34(11): 2079–2083.

15.

Mowry

Spyker

Brooks

, et al. 2015 Annual report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 33rd annual report. Clin Toxicol 2016; 54(10): 924–1109.

16.

Graudins

Lee

Druda

Calcium channel antagonist and beta-blocker overdose: antidotes and adjunct therapies. Br J Clin Pharmacol 2015; 81(3): 453–461.

17.

Chan

Tse

Lau

FL.

Hong Kong Poison Information Centre: annual report 2014. Hong Kong J Emerg Med 2015; 22(6): 376–387.

18.

Paladino

Subramanian

Nabors

, et al. The utility of shock index in differentiating major from minor injury. Eur J Emerg Med 2011; 18(2): 94–98.

19.

Obuchowski

Mcclish

DK.

Sample size determination for diagnostic accuracy studies involving binormal ROC curve indices. Stat Med 1997; 16(13): 1529–1542.

20.

Cannon

Braxton

Kling-smith

, et al. Utility of the shock index in predicting mortality in traumatically injured patients. J Trauma 2009; 67(6): 1426–1430.

21.

Ozawa

Hayashi

Kobori

New generation calcium channel blockers in hypertensive treatment. Curr Hypertens Rev 2006; 2(2): 103–111.