Sage Journals: Discover world-class research

Abstract

Background

Accurate assessment of disturbed consciousness level (DCL) is crucial for predicting acutely poisoned patients’ outcomes.

Aim

Development of a novel Poisoning Agitation-Sedation Score (PASS) to predict the need for endotracheal intubation (ETI) and mechanical ventilation (MV) in acutely poisoned patients with DCL. Validation of the proposed score on a new set of acutely poisoned patients with DCL.

Methods

This study was conducted on 187 acutely poisoned patients with DCL admitted to hospital from June 2020 to November 2021 (Derivation cohort). Patients' demographics, toxicological data, neurological examination, calculation of the Glasgow Coma Scale (GCS), Full Outline of Unresponsiveness (FOUR) score, Richmond Agitation–Sedation Scale (RASS), and outcomes were gathered for developing a new score. The proposed score was externally validated on 100 acutely poisoned patients with DCL (Validation cohort).

Results

The PASS assessing sedation consists of FOUR (reflexes and respiration) and GCS (motor) and provides a significantly excellent predictive power (AUC = 0.975) at a cutoff ≤9 with 100% sensitivity and 92.11% specificity for predicting the need for ETI and MV in sedated patients. Additionally, adding RASS (agitation) to the previous model exhibits significantly good predictive power (AUC = 0.893), 90.32% sensitivity, and 73.68% specificity at a cutoff ≤14 for predicting the need for ETI and MV in disturbed consciousness patients with agitation.

Conclusion

The proposed PASS could be an excellent, valid and feasible tool to predict the need for ETI and MV in acutely poisoned disturbed consciousness patients with or without agitation.

Keywords

Disturbed consciousness agitation sedation poisoning scores external validation

Introduction

Disturbed consciousness is a common major complaint in the emergency department. Many pathological etiologies of altered consciousness include sepsis, epilepsy, stroke, post-anoxic coma, organ failure, and metabolic disturbances.¹ Additionally, toxins can also cause altered consciousness, either directly through their action on the central nervous system or indirectly through inducing metabolic abnormalities (e.g., hypoglycemia or hyperammonemia caused by drug overdose).²

Impaired consciousness has many nonspecific terms that are often misused, leading to miscommunications. It may range from drowsiness, somnolence, confusion, delirium, lethargy, obtundation, and stupor to deep coma.³ Accordingly, scales have been designed to avoid subjective evaluations by quantitative consciousness assessment.⁴ Furthermore, scoring systems facilitate communication among examiners for monitoring consciousness changes, rapid therapeutic decisions, and outcomes prediction.⁵

Glasgow Coma Scale (GCS) has provided a practical tool for bedside assessment of consciousness level using the three components: verbal, motor, and eye response.⁶ Yet, it has some limitations in evaluating the verbal response in aphasic, intubated patients and inconsistency of inter-observer reliability.^7,8

Full Outline of Unresponsiveness (FOUR) score was developed to overcome the GCS confines with an accurate, comprehensive assessment of the patient’s neurological condition.⁹ Additionally, it evaluates respiratory patterns and brainstem reflexes that GCS does not assess.¹⁰ However, the FOUR score is more complicated, and its application remains quite difficult..¹¹ Additionally, it can not be used to assess disturbed consciousness patients with agitations.

The Richmond Agitation-Sedation Scale (RASS) is a simple tool that assesses agitation and sedation.¹² It can evaluate agitation status in patients whose consciousness level assessment by traditional scales is relatively complex.^13,14 Nevertheless, the RASS cannot evaluate consciousness in patients with visual or auditory impairments, as it relies on the presence of eye contact and vocal responses to verbal stimuli.¹⁵

Altered consciousness patients are at high risk for mortality and various morbidities such as aspiration pneumonia, hypoxia, persistent cognitive impairment, need for mechanical ventilation (MV), intensive care unit (ICU) admission, and prolonged hospitalization period.^16,17 Therefore, outcome prediction is a major concern for clinical toxicologists for early triaging of high-risk, acutely poisoned patients with disturbed consciousness, especially in developing countries with restricted hospital resources.^18,19

Although various scoring systems and models are used in acutely poisoned patients with disturbed consciousness to predict the need for MV and ICU admission, their predictive accuracy is still questionable owing to their limitations and inadequate clinical validation.²⁰^–²³ As a result, the traditional prognostic scores are frequently inaccurate for determining prognosis in acutely poisoned patients with an unexpected course that is dependent on numerous variables.^18,24

From this perspective, we aimed to develop and validate a novel Poisoning Agitation-Sedation Score (PASS) for predicting the need for endotracheal intubation (ETI) and mechanical ventilation (MV) in acutely poisoned patients with DCL.

Patients and methods

Study design and setting

This prospective cohort study was initially carried out on 187 acutely poisoned patients with disturbed consciousness who were admitted to Tanta University Poison Control Center (TUPCC) from the start of June 2020 till the end of November 2021 (the derivation cohort). Additionally, the novel proposed score for conscious assessment was validated on a new set of 100 disturbed conscious patients from the start of December 2021 to the end of May 2022 (the validation cohort).

Ethical considerations

This study was conducted following the approval of the medical research ethics committee of Tanta University Faculty of Medicine (33,809/5/20). Written informed consent was taken from patients’ legal guardians. The confidentiality of patients’ records was considered by keeping them anonymous.

Sample size

The sample size of the derivation cohort was calculated using G*Power version 3.1.9.4. (https://www.gpower.hhu.de/) based on the need for ETI and MV as the primary outcome. According to Mohammed and Gawesh,²⁵ 34% of 100 acutely poisoned patients with disturbed consciousness needed ETI and MV. The following assumptions were considered: two-tailed analysis, 90% power level, 5% margin of error, and an allocation ratio of 4. The total required sample size was 165. Additionally, 13% of the calculated sample size was added to compensate for attrition during the follow-up of patients. So, the finally required sample size was 187 patients.

Inclusion criteria

Figure 1 shows a flowchart of the eligibility criteria of patients included in our study. All admitted acutely poisoned patients with disturbed consciousness aged over 18 years (male or female) were included. The diagnosis was based on the history of any drug overdose or toxin exposure, characteristics of toxicological symptoms and signs, and available routine & toxicological investigations.

Figure 1.

Flow chart of the eligibility criteria of the studied patients.

Exclusion criteria

• Admitted patients more than 24 h after poisoning.

• Patients with previous neurological diseases that affect consciousness level.

• The presence of concomitant head trauma.

• Patients with chronic illnesses such as liver, kidney, and heart diseases

• Referred patients with previous medical intervention.

• Patients with missing data.

Data collection tool for the derivation and validation cohort

Patient characteristics

Demographic data included age and sex. Toxicological data included poison name, exposure route, and poisoning mode. On admission, vital signs and neurological examinations were recorded.

Laboratory investigations on admission

Arterial blood gases, serum electrolyte levels, random blood sugar levels, liver enzymes, urea, serum creatinine, and complete blood count were done. Toxicological tests were performed, including a silver nitrate test to confirm phosphide exposure and serum pseudocholinesterase level to identify organophosphorus exposure. Qualitative urine screening was performed to detect cannabinoids, tramadol, opiates, benzodiazepines, and barbiturates using ACON® DOATM kits (one-step Drug of Abuse rapid test).²⁶

Scores calculation for the derivation cohort on admission

The GCS assesses patients according to three aspects of responsiveness (eye-opening, motor, and verbal responses.). It ranges from 3 (deep coma) to 15 (fully awake person). GCS grades are mild (13-14), moderate (9-12), and severe (less than 9).^27,28

The FOUR score includes four components assessing neurological function: motor response, brainstem reflexes, eye response, and respiratory patterns. Each item is given a score ranging from 0 to 4. The total score ranges from 0 to 16. The lower score values indicated a lower level of consciousness.⁹

The RASS is a 10-point scale ranging from −5 to +4. It has five levels of sedation ranging from 1 (drowsy) to −5 (unarousable), four levels of agitation that range from +1 (restless) to +4 (combative), and level 0 is “calm and alert”.²⁹

Patient outcomes and treatment

Patients were categorized according to the need for ETI and MV into patients who needed ETI and MV and did not. All patients were treated according to the TUPCC protocol that follows the Shannon³⁰; Olson³¹; Leikin and Carlson³²; Nelson et al.³³ guidelines for treating acutely poisoned patients.

• The patients in the present study were intubated and mechanically ventilated according to the following criteria reported by Nolan and Kelly; Brown et al.^34,35:

➢ Respiratory insufficiency (respiratory rate < 8 or > 30 per minute, PO2 in blood gas < 55 mmHg, PCO 2 > 55 mmHg.

➢ Non-compensated acidosis-alkalosis.

➢ Upper and lower airway obstruction.

➢ Persistent hemodynamic instability.

➢ Excessive chest secretions with a high risk of aspiration.

Statistical methods

All data were analyzed by the Statistical Package for the Social Sciences (SPSS) version 26 (IBM Corp., Armonk, NY, USA). Categorical variables were summarized as frequencies and percentages. Comparison of the categorical variables was performed by the Independent Chi-Square or Fisher’s Exact tests as appropriate. Numerical variables were checked for distribution by the Shapiro-Wilk test, skewness, and kurtosis and presented with median and interquartile range (25th–75th percentile). The difference between the two groups and the associations between the components of the scores were performed by the Mann-Whitney U test. Consequently, two multivariable stepwise logistic regression analyses were performed to develop PASS for predicting the need for ETI and MV. The first was the model that involved disturbed consciousness patients without agitation (n = 113); the entered components in the regression analysis were eye & motor components of GCS, all components of FOUR, and RASS sedation. The second was the model for predicting the need for ETI and MV in disturbed consciousness patients with agitation (n = 74) that involved eye ‘& motor components of GCS, all components of FOUR score, and RASS (agitation).

The model’s performance was evaluated by the concordance (C) index as a discrimination measurement tool. The C index measures how well the model discriminates between patients who needed and did not need ETI and MV and can be viewed as the extension of the area under the receiver operating characteristics (ROC) curve. It varies theoretically between 0.5 and 1.0, with a value of 1.0 indicating perfect discrimination. p value <.05 was considered statistically significant. These two prognostic models were externally validated in a cohort of 100 patients. The ROC curve assessed the performance of the proposed score to predict the need for ETI and MV in the validation cohort. The area under the curve (AUC) is graded as follows: 0.6–0.7 = poor, 0.7–0.8 = fair, 0.8–0.9 = good, and 0.9–1 = excellent or perfect discrimination.

Results

The median age of the studied patients was 25 years in both groups. Table 1 demonstrated that males represented 52.9% of the derivation and 58% of the validation cohorts. Most ingestions were suicidal (84.5% and 92%) by oral route (87.7% and 94%) in the derivation and validation cohorts, respectively. Moderate GCS and agitations were commonly observed in both groups with no significant differences (p = .333 and >0.089, respectively). Sixty-one patients (32.6%) and forty-three patients (43%) needed ETI and MV in the derivation and validation cohorts, respectively. There were no significant statistical differences between the derivation and the validation cohorts regarding demographics & toxicological data, neurological examination, and the need for ETI and MV (p > .05). Although there was a significant difference between groups regarding poison types, the most frequent poisons included in the derivation and validation cohorts were antipsychotics (29.4% and 39%, respectively) and aluminum phosphide (13.9% and 26%, respectively). Additionally, no significant statistical differences were observed between the derivation and the validation cohort as regards clinical examinations and laboratory investigations (p > .5), as shown in supplementary table 1.

Table 1.

Patients characteristics in the derivation and validation cohort.

Variables		Derivation cohort (n = 187)		Validation cohort (n = 100)		p-value
Age, years	Minimum-maximum	18–70		18–79		0.346^a
Age, years	Median (IQR)	25 (20–33)		25 (20–38)		0.346^a
Sex	Male	99	52.9%	58	58%	0.412^b
Sex	Female	88	47.1%	42	42%	0.412^b
Poison type	Antipsychotics	55	29.4%	39	39%	0.003*^b
	Aluminum phosphide	26	13.9%	26	26%
	Benzodiazepines	18	9.6%	7	7%
	Antiepileptics	18	9.6%	4	4%
	Heroin	17	9.1%	3	3%
	Organophosphates	10	5.3%	9	9%
	Antidepressants	16	8.6%	2	2%
	Carbon monoxide	4	2.1%	3	3%
	Methanol	3	1.6%	3	3%
	Baclofen	6	3.2%	0	0%
	Tramadol	3	1.6%	2	2%
	Ethanol	3	1.6%	1	1%
	Co-ingestion	4	2.1%	0	0%
	Carbamate	3	1.6%	0	0%
	Dormex	1	0.5%	1	1%
Route of exposure	Oral	164	87.7%	94	94%	0.33^b
	Injection	13	7%	3	3%
	Inhalation	7	3.7%	3	3%
	Combinedoral and inhalation	3	1.6%	0	0%
Alleged mode of poisoning	Suicidal	158	84.5%	92	92%	0.111^b
	Abuse	24	12.8%	5	5%
	Accidental	5	2.7%	3	3%
Consciousness level according to GCS	Mild	54	28.9%	26	26%	0.333^b
	Moderate	75	40.1%	49	49%
	Severe	58	31%	25	25%
Mental state according to RASS	Agitation	113	60.4%	50	50%	0.089^b
Mental state according to RASS	Sedation	74	39.6%	50	50%	0.089^b
Need for endotracheal intubation and mechanical ventilation	No	126	67.4%	57	57%	0.081^b
Need for endotracheal intubation and mechanical ventilation	Yes	61	32.6%	43	43%	0.081^b

n: Number, %: Percentage, IQR: Interquartile range, GCS: Glasgow Coma Scale, RASS: Richmond Agitation–Sedation Scale, Tests of significance; *Significant at p < .05.

^aMann-Whitney test.

^bChi square test.

The medians (IQR) and/or mean ranks of GCS, FOUR, and RAAS (sedation) components were significantly lower in patients who needed ETI and MV than their counterparts. Conversely, medians of RASS (agitation) were significantly higher in patients who needed ETI and MV than in those who did not need it (p < .001), as illustrated in Table 2.

Table 2.

Comparison between patients in the derivation cohort who needed endotracheal intubation and mechanical ventilation and those who did not need as regards the GCS, FOUR, and RASS scores components.

Scores	Scores components		Need for endotracheal intubation and mechanical ventilation		Total (n = 187)	Mann-whitney test
Scores	Scores components		No (n = 126)	Yes (n = 61)	Total (n = 187)	p –value
GCS	Eye	Median	3	2	3	<0.001*
		IQR	2–4	1–3	2–4
		Mean rank	105.12	71.03
	Motor	Median	5	5	5	0.004*
		IQR	5–5	1–5	5–5
		Mean rank	100.81	79.93
	Verbal	Median	3	3	3	0.551
		IQR	2–4	1–4	2–4
		Mean rank	95.6	90.7
FOUR	Eye	Median	2	1	2	<0.001*
		IQR	1–4	0–2	1–4
		Mean rank	104.23	72.86
	Motor	Median	3	3	3	0.006*
		IQR	3–3	0–3	3–3
		Mean rank	100.62	80.32
	Reflexes	Median	4	4	4	<0.001*
		IQR	4 – 4	2 – 4	4 – 4
		Mean rank	105.81	69.61
	Respiration	Median	4	2	4	<.001*
		IQR	4–4	2–2	2–4
		Mean rank	115.85	48.86
RASS	Sedation	Median	−3	−5	−4	<0.001*
		IQR	−4 to −2	−5 to −4	−4 to −2
		Mean rank	67.33	28.19
	Agitation	Median	+2	+3	+2	<0.001*
		IQR	+1– +2	+2– +3	+1– +3
		Mean rank	27.88	51.62

IQR: Interquartile range, GCS: Glasgow Coma Scale, FOUR: Full Outline of Unresponsiveness, RASS: Richmond Agitation–Sedation Scale, *Significant at p < .05.

The ROC curve analysis revealed that the FOUR (respiration) and RASS (sedation and agitation) components had significantly good predictive power in predicting the need for ETI and MV in the derivation group (AUC = 0.858, 0.849, and 0.821, respectively). Furthermore, pairwise comparisons of AUCs of the studied components revealed significant differences between FOUR (respiration) and other components. However, GCS (eye and motor) and FOUR (eye, motor, and reflexes) components had poor significant predictive power in outcome discrimination, as shown in Table 3 and Figures 2–5.

Table 3.

Diagnostic performance of the GCS, FOUR, and RASS scores components for predicting the need for endotracheal intubation and mechanical ventilation in the derivation cohort.

Scores Components	GCS		Four				RASS
Scores Components	Eye	Motor	Eye	Motor	Reflexes	Respiration	Sedation	Agitation
AUC	0.682	0.612	0.668	0.609	0.694	0.858	0.849	0.821
p-value	<0.001*	0.014*	<.001*	0.017*	<0.001*	<0.001*	<0.001*	<0.001*
AUC 95%CI	0.61–0.748	0.538–0.682	0.595–0.735	0.535–0.679	0.622–0.759	0.8–0.905	0.77–0.91	0.714–0.9
Cutoff	≤1	≤3	≤0	≤2	≤2	≤2	≤−4	>+2
Sensitivity%	44.26%	36.07%	42.62%	36.1%	40.98%	81.97%	96.77%	66.67%
Specificity%	87.3%	92.86%	87.3%	92.9%	97.62%	89.68%	58.02%	84.09%
Accuracy%	73.26%	74.33%	65%	74.3%	79.1%	87.2%	81.3%	77.03%
p-value from pairwise comparisons of AUCs
Respiration	0.002*	0.001*	0.001*	0.001*	0.004*		0.005*	0.004*

%: percentage AUC: area under the curve, CI: Confidence interval, GCS: Glasgow Coma Scale, FOUR: Full Outline of Unresponsiveness, RASS: Richmond Agitation–Sedation Scale, *Significant at p < .05.

Figure 2.

ROC curves of eye and motor component of GCS for predicting the need for endotracheal intubation and mechanical ventilation in the derivation cohort.

Figure 3.

ROC curves of FOUR score components for predicting the need for endotracheal intubation and mechanical ventilation in the derivation cohort.

Figure 4.

ROC curve of RASS (sedation) for predicting the need for endotracheal intubation and mechanical ventilation in the derivation cohort.

Figure 5.

ROC curve of RASS (agitation) for predicting the need for endotracheal intubation and mechanical ventilation in the derivation cohort.

To develop a PASS for predicting the need for ETI and MV, the studied patients in the derivation cohort were categorized according to the presence or absence of agitation into two categories: patients without agitation (113) and patients with agitation (74). Consequently, significant variables were entered in multivariable stepwise logistic regression analysis.

The logistic regression analysis showed that the model assessing sedation, including FOUR (reflex and respiration) and GCS (motor) components, had excellent predictive power for ETI and MV in disturbed consciousness patients (AUC = 0.972, 93.55% sensitivity, and 95.12% specificity) with an overall accuracy of 94.69%. Additionally, adding the agitation component of RASS to the previous model had excellent discrimination power for predicting the need for ETI and MV in disturbed consciousness patients with agitation (AUC = 0.946%, 86.67% sensitivity, and 86.36% specificity) with an accuracy of 86.49% (Table 4).

Table 4.

Multivariable logistic regression analysis (stepwise) prognostic model for predicting the need for endotracheal intubation and mechanical ventilation in disturbed consciousness patients with or without agitation.

Variables	Beta coefficient	p-value	AOR	95% CI AOR	Hosmer&Lemeshow test	Sensitivity%	Specificity%	Accuracy%	AUC	R2%	p-value
Patients without agitation (n = 113)
Reflexes (FOUR)	−1.758	0.007*	0.172	0.048–0.617	0.895	93.55%	95.12%	94.69%	0.972 (95% CI: 0.922–0.994	82.76%	<0.001*
Respiration (FOUR)	−2.206	<0.001*	0.11	0.041–0.292
Motor (GCS)	−0.654	0.001*	0.519	0.313–0.862
Patients with agitation (n = 74)
Reflexes (FOUR)	2.956	0.032*	0.052	0.003–0 .784	0.748	86.67%	86.36%	86.49%	0.946 (95% CI: 0.868–0.985)	72.64	<0.001*
Respiration (FOUR)	1.551	<0.001*	0.211	0.091 −0.49
Motor (GCS)	1.691	0.029*	5.429	1.183 −24.926
RASS (agitation)	1.174	0.041*	3.237	1.045 −10.03

AOR: Adjusted odds ratio, %: percentage, AUC: area under the curve CI: Confidence interval, GCS: Glasgow Coma Scale, FOUR: Full Outline of Unresponsiveness, RASS: Richmond Agitation–Sedation Scale, *Significant at p < .05.

The proposed PASS assessing sedation in acutely poisoned patients with disturbed consciousness includes reflexes & respiration components of FOUR, and motor of GCS ranging from 1 to 14. Meanwhile, the PASS assessing agitation in acutely poisoned patients with disturbed consciousness consists of reflexes & respiration of FOUR, the motor of GCS, and RASS (agitation) ranging from 2 to 18, as demonstrated in Table 5.

Table 5.

Components of the proposed Poisoning Agitation-Sedation Score (PASS) for predicting the need for endotracheal intubation and mechanical ventilation in acutely poisoned patients with disturbed consciousness with or without agitation.

Parameters	Response	Grade
Reflexes	Pupil and corneal reflexes present	4
	One pupil wide and fixed	3
	Pupil or corneal reflexes absent	2
	Pupil and corneal reflexes absent	1
	Absent pupil, corneal, and cough reflex	0
Respiration	Not intubated, regular breathing pattern	4
	Not intubated, Cheyne–Stokes breathing pattern	3
	Not intubated, irregular breathing	2
	Breathes above ventilator rate	1
	Apnea or breathes at ventilator rate	0
Motor	Obeys commands	6
	Localizes pain	5
	Flexion withdrawal	4
	Decerebrate flexion	3
	Decerebrate extension	2
	No response	1
Agitation	Overtly combative or violent; immediate danger to staff	4
	Pulls on or removes tube(s) or catheter(s) or has aggressive behaviour toward staff	3
	Frequent non purposeful movement or patient–ventilator dyssynchrony	2
	Anxious or apprehensive but movements not aggressive or vigorous	1

By comparing the predictive power of PASS with other existing scales in the validation cohort, PASS sedation and PASS agitation had a significantly excellent and good discriminatory power (AUC = 0.975 and 0.893, respectively) and both had higher overall accuracy (96% and 82%, respectively) than GCS, FOUR and RASS scores as illustrated in Table 6 and Figures 6 and 7.

Table 6.

Performance of GCS, FOUR, RASS, and PASS scores in predicting the need for endotracheal intubation and mechanical ventilation in the validation cohort.

Variables		Cutoff	Sensitivity%	Specificity%	Accuracy%	AUC	AUC 95% CI	p value
Scores for disturbed consciousness without agitation	GCS	≤9	75.00	52.63	76.0	0.582	0.434 to 0.720	0.389
	FOUR	≤12	75.00	39.47	76.0	0.547	0.400 to 0.688	0.640
	RASS	≤−4	75.00	39.47	76.0	0.561	0.414 to 0.701	0.517
	PASS	≤9	100.00	92.11	96.0	0.975	0.885 to 0.999	<0.001*
	Pairwise comparison of ROC curves (difference between AUCs) revealed the following significant difference (p < .001*)
	PASS versus GCS p < .001*
	PASS versus FOUR p = .006*
	PASS versus RASS p < .001*

Variables		Cutoff	Sensitivity %	Specificity %	Accuracy %	AUC	AUC 95% CI	p value
Scores for disturbed consciousness with agitation	GCS	≤12	67.74	15.79	62.0	0.507	0.362 to 0.651	0.932
	FOUR	≤11	70.97	52.63	60.0	0.612	0.464 to 0.747	0.173
	RASS	>2	67.44	78.95	62.0	0.509	0.364 to 0.653	0.903
	PASS	≤14	90.32	73.68	82.0	0.893	0.773 to 0.963	<0.001*
	Pairwise comparison of ROC curves (difference between AUCs) revealed the following significant difference (p < .001*)
	PASS versus GCS p < .001*
	PASS versus FOUR p < .001*
	PASS versus RASS p < .001*

%: Percentage AUC: Area under the curve, CI: Confidence interval, GCS: Glasgow Coma Scale, FOUR: Full Outline of Unresponsiveness, RASS: Richmond Agitation–Sedation Scale, PASS: Poisoning Agitation-Sedation Score, ROC: receiver operating characteristics, *Significant at p < .05.

Figure 6.

ROC curves of GCS, FOUR, RASS, and PASS (sedation) for predicting the need for endotracheal intubation and mechanical ventilation in the validation cohort.

Figure 7.

ROC curves of GCS, FOUR, RASS, and PASS (agitation) for predicting the need for endotracheal intubation and mechanical ventilation in the validation cohort.

Discussion

The decision for ETI and MV is challenging, especially in high-risk, acutely poisoned patients with disturbed consciousness.³⁶ Additionally, there is an urgent need to develop a practical score accurately assessing the level of consciousness in disturbed conscious patients to improve healthcare professionals' communication and patient outcomes.

The current study revealed that the median age of the studied patients in both groups was 25 years, with male predominance (52.9% and 58%) in the derivation and validation cohorts, respectively. Similarly, Morsy et al.³⁷ recorded that the median age of acutely poisoned patients with DCL admitted to TUPCC was 23 years, with males outnumbered females (56.8% vs 53.2%). The high incidence of poisoning in young males may be attributed to impulsive behavior, family conflict, peer pressure, conflict with laws, unemployment, failure in love, and professional frustration.³⁸

The most frequently involved poisons in both groups of the present study were antipsychotics, followed by aluminum phosphide administrated for suicidal intention. Similarly, El-Sarnagawy et al.³⁹ demonstrated that antipsychotic drugs and phosphides were the most causative pharmaceutical and non-pharmaceutical substances in acutely poisoned patients. The availability of antipsychotic medications and uncontrolled sales of aluminum phosphide could explain the higher incidence of these substances’ poisoning.^40,41 Furthermore, aluminum phosphide is a commonly used pesticide in our locality due to the agricultural nature of the Gharbia governorate.⁴²

Regarding GCS assessment, the highest percentage of patients presented with moderate GCS (40.1% and 49%) in both the derivation and validation cohorts, respectively. However, Zaghary et al.⁴³ reported that 9% of acutely poisoned patients admitted to Sohag University Hospitals belonged to the moderate group. The difference in the type of poisons, grading severity, and the consumed amount could explain variations among the GCS results of these studies.⁴⁴ Agitation was reported in 39% and 50% of the studied patients in the derivation and validation cohorts, respectively. Agitation result from either direct enhancing transmission of excitatory amino acids or suppression of the inhibitory GABA neurons or secondary events, such as hypoxia.⁴⁵

The current study revealed that 32.6% of patients in the derivation cohort needed ETI and MV compared to 43% of patients in the validation cohort. Near similar results, El-Sarnagawy and Hafez,²³ in their studies on acutely poisoned patients with disturbed consciousness levels admitted to TUPCC, found that 23% of their patients needed ETI and MV.

Comparison between the components of GCS, FOUR, and RASS scores regarding predicting the need for ETI and MV in the derivation cohort revealed that both eye (≤1) and motor (≤3) components of GCS were significantly valid in predicting the need for ETI and MV with poor prediction power (AUC = 0.682 and 0.612, respectively). On the contrary, Mood et al.⁴⁴ disclosed good and excellent discriminative power of motor and eye components of GCS (AUC = 0.89 and 0.911, at cutoff ≤2 and ≤5, respectively) in predicting complications in acute mixed drug poisoned patients. Furthermore, Beskind et al.⁴⁶ confirmed the excellent discriminatory power of the motor component of GCS (AUC = 0.95) for predicting the need for ETI in disturbed consciousness patients after head trauma.

Concerning the FOUR score, the respiration component had a good prediction power (AUC = 0.858, at a cutoff of ≤2) for the likelihood of ETI and MV requirements. Alternatively, eye, motor, and reflex components were significantly valid in predicting the need for ETI and MV with poor discriminatory power (AUC = 0.668, 0.609, and 0.694, respectively). Wijdicks et al.⁴⁷ stated that respiration and brainstem reflexes assessment improve the predictive power as brainstem damage and its failure to maintain adequate respiration reflect poisoning severity. Furthermore, brainstem reflexes provide a complete assessment of the coma depth.¹⁰ Besides, including the respiratory component of FOUR could determine irregular breathing patterns and Cheyne-stokes breathing that indicate lower brainstem dysfunction of respiratory control. Conversly to our finding, Okasha et al.⁴⁸ reported that the discriminative power was excelent for the eye component (AUC = 0.944), good for the motor component (AUC = 0.820), and poor for both brainstem and respiration components (AUC = 0.681, AUC = 0.617, respectively) for predicting the need for ETI in disturbed conscuious patients after traumatic brain injury. This discrepancy in the dicriminatory power of FOUR score components may be due to different patient populations and variable severity of illness.

Our patients who required ETI and MV, had a significantly lower median RASS (sedation) than those who did not need (−5 vs −3, respectively). In line with that, Zadravecz et al.⁴⁹ found that patients admitted in the general wards with negative RASS scores had poor outcomes. Moreover, Ko et al.⁵⁰ reported that deeply sedated patients (score>−3) were associated with increased in-hospital mortality in patients admitted to the cardiac ICU. Additionally, the median of RASS (agitation) was significantly increased in patients who needed ETI and MV than those who did not need (+3 vs + 2, respectively). It had good predictive power for the need of ETI and MV (AUC = 0.821). Likewise, Sabzghabaee et al.⁵¹ noted that agitations were observed in acute poisoned patients following illegal substance abuse with mean RASS score +3. Furthermore, they concluded that agitation increases the risk of complications.

Few studies compared scoring systems for predicting the need for ETI and MV in acutely poisoned patients with drug overdose, pesticides, tricyclic antidepressants (TCAs), carbon monoxide (CO), and clozapine.^23–52 However, previously compared scores cannot be applied to all poisoned patients with disturbed consciousness. Additionally, they cannot assess the risk of agitated patients needing ETI and MV.

From this perspective, we proposed a novel PASS to predict the need for ETI and MV in all acutely poisoned patients with disturbed consciousness, either with or without agitation. The PASS assessing sedation consists of three variables (reflexes & respiration components of FOUR and motor component of GCS). Meanwhile, PASS assessing agitation includes adding RASS (agitation) to the previous score to predict the need for ETI and MV in acutely poisoned patients with disturbed consciousness with agitation. The accuracy proposed PASS was externally validated on a new set of 100 patients (the validation cohort). The PASS assessing sedation in disturbed consciousness patients without agitation had a significantly excellent predictive power (AUC = 0.975) at a cut off ≤9 with 100% sensitivity, 92.11% specificity, and total accuracy of 96% for predicting the need for ETI and MV. Additionally, the PASS assessing agitations in disturbed consciousness patients with the agitation was significantly valid in predicting the need for ETI and MV with good prediction power (AUC = 0.893), 90.32% sensitivity, and 73.68% specificity at a cutoff of ≤14.⁵³^–⁵⁵

The PASS score has advantages over the conventional scores, making it more suitable for predicting the need for ETI and MV as it assesses the breathing pattern, which is essential for predicting the need for assisted ventilation in disturbed-conscious patients. Additionally, the respiratory component of FOUR enables clinicians to assess the breathing patterns of apneic and irregular breathing patients. Moreover, the brain stem component provides valuable information about brain stem affection.^56,57 Additionally, the motor component is the most reliable component of GCS for predicting outcomes as it has a good inter-observer agreement and can be easily assessed, even in intubated, inebriated, or sedated patients, as well as, it is the most affected component in severely injured patients.⁵⁸^–⁶⁰ Furthermore, adding RASS agitation to the PASS model allows for better evaluation of disturbed consciousness patients with agitation.

Strength and limitation

The proposed PASS score has many potential implications in clinical practice due to its validity and easy applicability for rapid neurological assessment, especially in settings with limited resources. Although the PASS score was externally validated, the studied patients were relatively small as it is a single-center study, which may not represent all poisoned patients.

Conclusion and recommendations

From the existing data, the proposed PASS is a simple clinical valid score providing an excellent discrimination power for predicting the need for ETI and MV in acutely poisoned disturbed consciousness patients with or without agitation. Accordingly, applying on-admission PASS assessment in toxicological centers may improve the patient's outcomes with reasonable healthcare resource utilization.

Further studies should be implemented on a wide scale with a larger sample of patients to assess the PASS's discriminatory ability for evaluating potential outcomes. Similarly, exploring the PASS applicability in various clinical settings or populations is essential to generalize its utilization. Additionally, investigating the long-term outcomes of patients assessed with PASS is recommended to endorse its validity and relevance. Furthermore, combining PASS with other clinical or laboratory markers is suggested to enhance its objectivity and predictive accuracy.

Supplemental Material

Supplemental Material - Development and validation of a novel poisoning agitation-sedation score for predicting the need for endotracheal intubation and mechanical ventilation in acutely poisoned patients with disturbed consciousness

Supplemental Material for Development and validation of a novel poisoning agitation-sedation score for predicting the need for endotracheal intubation and mechanical ventilation in acutely poisoned patients with disturbed consciousness by Soha A. Abd Elghany, Heba I. Lashin, Ghada N. El-Sarnagawy, Merfat M. Oreby, and Eman Soliman in Human & Experimental Toxicology.

Footnotes

Acknowledgments

The authors extend their appreciation to Tanat University Poison Control Center (TUPCC) for facilitating data collection.

Author contributions

Soha A. Abd Elghany contributed to the study conception, data collection, methodology, analysis, interpretation, and manuscript writing.; Heba I. Lashin contributed to the study conception, data collection, methodology, analysis, interpretation, and manuscript writing; Ghada N. El-Sarnagawy: contributed to the study conception, data collection, methodology, analysis, interpretation, and manuscript writing; Merfat M. Oreby; contributed to the study conception, data collection, methodology, analysis, interpretation, and manuscript writing; Eman Soliman; contributed to the study conception, data collection, methodology, analysis, interpretation, and manuscript writing.

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.

ORCID iDs

Soha A. Abd Elghany

Heba I. Lashin

Ghada N. El-Sarnagawy

Data Availability Statement

Data are available upon reasonable request from the corresponding author.

Supplemental Material

Supplemental material for this article is available online.

References

Horsting

Franken

Meulenbelt

, et al. The etiology and outcome of non-traumatic coma in critical care: a systematic review. BMC Anesthesiol 2015; 15: 65–70. DOI: 10.1186/s12871-015-0041-9.

Song

Wang

. Altered level of consciousness: evidence-based management in the emergency department. Pediatr Emerg Med Pract 2017; 14: 1–28.

Kalita

Misra

. In search of a better measuring scale of consciousness. Ann Card Anaesth 2019; 22: 149–150. DOI: 10.4103/aca.ACA_193_18.

Wang

Liu

, et al. Study of simplified coma scales: acute stroke patients with tracheal intubation. Chin Med J (Engl) 2018; 131: 2152–2157. DOI: 10.4103/0366-6999.240813.

Kornbluth

Bhardwaj

. Evaluation of coma: a critical appraisal of popular scoring systems. Neurocritical Care 2011; 14: 134–143. DOI: 10.1007/s12028-010-9409-3.

Teasdale

Maas

Lecky

, et al. The Glasgow Coma Scale at 40 years: standing the test of time. Lancet Neurol 2014; 13: 844–854. DOI: 10.1016/s1474-4422(14)70120-6.

Mercy

Thakur

Yaddanpudi

, et al. Can FOUR Score replace GCS for assessing neurological status of critically ill patients-An Indian Study. Nurs Midwifery Res J 2013; 9: 63–72. DOI: 10.33698/NRF0154.

Saika

Bansal

Philip

, et al. Prognostic value of FOUR and GCS scores in determining mortality in patients with traumatic brain injury. Acta Neurochir 2015; 157: 1323–1328. DOI: 10.1007/s00701-015-2469-6.

Wijdicks

Bamlet

Maramattom

, et al. Validation of a new coma scale: the FOUR score. Ann Neurol 2005; 58: 585–593. DOI: 10.1002/ana.20611.

10.

Iyer

Mandrekar

Danielson

, et al. Validity of the FOUR score coma scale in the medical intensive care unit. Mayo Clin Proc 2009; 84: 694–701. DOI: 10.1016/s0025-6196(11)60519-3.

11.

Anestis

Tsitsopoulos

Tsonidis

, et al. The current significance of the FOUR score: a systematic review and critical analysis of the literature. J Neurol Sci 2020; 409: 116600–116620. DOI: 10.1016/j.jns.2019.116600.

12.

Kerson

DeMaria

Mauer

, et al. Validity of the Richmond agitation-sedation scale (RASS) in critically ill children. J Intensive Care 2016; 4(1): 65–66. DOI: 10.1186/s40560-016-0189-5.

13.

Mégarbane

. Toxidrome-based approach to common poisonings. APJMT 2014; 3(1): 2–12. DOI: 10.22038/apjmt.2014.2463.

14.

Rajabi Kheirabadi

Tabeshpour

Afshari

. Comparison of three consciousness assessment scales in poisoned patients and recommendation of a new scale: AVPU plus. APJMT 2015; 4: 58–63. DOI: 10.22038/apjmt.2015.5080.

15.

Kerson

DeMaria

Mauer

, et al. Validity of the Richmond agitation-sedation scale (RASS) in critically ill children. J Intensive Care 2016; 4: 65–70. DOI: 10.1186/s40560-016-0189-5.

16.

Lauterbach

Voss

Gerigk

, et al. Bacteriology of aspiration pneumonia in patients with acute coma. Intern Emerg Med 2014; 9: 879–885. DOI: 10.1007/s11739-014-1120-5.

17.

Mehta

Cook

Devlin

SLEAP Investigators Canadian Critical Care Trials Group , et al.. Prevalence, risk factors, and outcomes of delirium in mechanically ventilated adults. Crit Care Med 2015; 43: 557–566. DOI: 10.1097/ccm.0000000000000727.

18.

Alizadeh

Hassanian-Moghaddam

Shadnia

, et al. Simplified acute physiology score II/acute physiology and chronic health evaluation II and prediction of the mortality and later development of complications in poisoned patients admitted to intensive care unit. Basic Clin Pharmacol Toxicol 2014; 115: 297–300. DOI: 10.1111/bcpt.12210.

19.

Shadnia

Zamani

Hassanian-Moghaddam

, et al. Prognostic value of cortisol and thyroid function tests in poisoned patients admitted to toxicology ICU. World J Emerg Med 2018; 9: 51–55. DOI: 10.5847/wjem.j.1920-8642.2018.01.008.

20.

Brandenburg

Brinkman

de Keizer

, et al. The need for ICU admission in intoxicated patients: a prediction model. Clin Toxicol 2017; 55: 4–11. DOI: 10.1080/15563650.2016.1222616.

21.

Shahin

Hafez

. Comparison of different scoring systems in poisoning with cholinesterase inhibitors. MJFMCT. 2020; 28(1): 25–42. doi:10.21608/MJFMCT.2020.19673.1005

22.

Elgazzar

Afifi

Shama

MAE

, et al. Development of a risk prediction nomogram for disposition of acute toxic exposure patients to intensive care unit. Basic Clin Pharmacol Toxicol 2021; 129: 256–267. DOI: 10.1111/bcpt.13619.

23.

El-Sarnagawy

Hafez

. Comparison of different scores as predictors of mechanical ventilation in drug overdose patients. Hum Exp Toxicol 2017; 36: 539–546. DOI: 10.1177/0960327116655389.

24.

Han

Kim

Lee

, et al. Development and validation of new poisoning mortality score system for patients with acute poisoning at the emergency department. Crit Care 2021; 25: 29–32. DOI: 10.1186/s13054-020-03408-1.

25.

Mohammed

Gawesh

E-S

. Acute poisoning induced coma: characteristics and predictive role of early creatine phosphokinase on its outcome. AJFM 2019; 32: 1–9. DOI: 10.21608/AJFM.2019.23376.

26.

Moeller

Lee

Kissack

. Urine drug screening: practical guide for clinicians. Mayo Clin Proc 2008; 83: 66–76. DOI: 10.4065/83.1.66.

27.

Teasdale

Jennett

. Assessment of coma and impaired consciousness. A practical scale. Lancet 1974; 2: 81–84. DOI: 10.1016/s0140-6736(74)91639-0.

28.

Middleton

. Practical use of the Glasgow Coma Scale; a comprehensive narrative review of GCS methodology. Australas Emerg Nurs J 2012; 15: 170–183. DOI: 10.1016/j.aenj.2012.06.002.

29.

Sessler

Gosnell

Grap

, et al. The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. Am J Respir Crit Care Med 2002; 166: 1338–1344. DOI: 10.1164/rccm.2107138.

30.

Shannon

. Emergency management of poisoning. In: Shannon

Borron

Burns

(eds). Haddad and Winchester's clinical management of poisoning and drug overdose. 4th ed.. Philadelphia: Saunders elsevier, 2007, pp. 13–61.

31.

Olson

. Comprehensive evaluation and treatment. In: Olson

Anderson

Benowitz

, et al. (eds). Poisoning & drug overdose. Chicago, IL: McGraw Hill Medical, 2012, pp. 1–61.

32.

Leikin

Carlson

. Toxicant-Induced alterations in consciousness. In: Brent

Burkhart

Dargan

, et al. (eds). Critical Care Toxicology. Cham: Springer, 2017, pp. 425–446.

33.

Nelson

Howland

Lewin

, et al. Principles of managing the acutely poisoned or overdosed patient. In: Hoffman

Howland

Lewin

, et al. (eds). Goldfrank's Toxicologic Emergencies. New York, NY: McGraw-Hill, 2019, pp. 33–41.

34.

Nolan

Kelly

. Airway challenges in critical care. Anaesthesia 2011; 66(2): 81–92. DOI: 10.1111/j.1365-2044.2011.06937.x.

35.

Brown

3rd Bair

Pallin

NEAR III Investigators , et al.. Techniques, success, and adverse events of emergency department adult intubations. Ann Emerg Med 2015; 65: 363–370. DOI: 10.1016/j.annemergmed.2014.10.036.

36.

Burket

Horowitz

Hendrickson

, et al. Endotracheal intubation in the pharmaceutical-poisoned patient: a narrative review of the literature. J Med Toxicol 2021; 17: 61–69. DOI: 10.1007/s13181-020-00779-3.

37.

Morsy

Wahdan

Abo El-Noor

, et al.. Assessment of acute toxicity cases with disturbed consciousness in Master Thesis. Tanta, Egypt: Tanta University Poison Control Centre, 2020.

38.

Sharma

Neelanjana

Rawat

, et al. Mortality and morbidity associated with acute poisoning cases in north-east India: a retrospective study. J Fam Med Prim Care 2019; 8: 2068–2072. DOI: 10.4103/jfmpc.jfmpc_237_19.

39.

El-Sarnagawy

Hafez

Amer

. Characteristics of suicidal poisoned patients admitted to tertiary care center during COVID-19 pandemic. Egypt J Neurol Psychiatr Neurosurg 2022; 58: 138–140. DOI: 10.1186/s41983-022-00577-4.

40.

Asawari

Atmaram

Bhagwan

, et al. Toxicological pattern of poisoning in urban hospitals of western India. J Young Pharm 2017; 9(3): 315–320. DOI: 10.5530/jyp.2017.9.63.

41.

Elagamy

Gabr

. Predictors of the need for Intensive Care Unit admission in acute organophosphorus poisoning: one year prospective study. Egypt J Forensic Sci Appl Toxicol 2019; 19: 1–9. DOI: 10.21608/ejfsat.2019.11316.1065.

42.

El-Sarnagawy

Abdelnoor

Abuelfadl

, et al. Comparison between various scoring systems in predicting the need for intensive care unit admission of acute pesticide-poisoned patients. Environ Sci Pollut Res Int 2022; 29: 33999–34009. DOI: 10.1007/s11356-021-17790-y.

43.

Zaghary

MMS

Radwan

Elsayed

. Validity of vital signs, coma scales and modified Apache score in prediction of prognosis and outcome of acutely poisoned patients. Egypt J Hosp Med 2021; 85(1): 2758–2766. DOI: 10.21608/EJHM.2021.189851.

44.

Eizadi Mood

Sabzghabaee

Khalili-Dehkordi

. Applicability of different scoring systems in outcome prediction of patients with mixed drug poisoning-induced coma. Indian J Anaesth 2011; 55: 599–604. DOI: 10.4103/0019-5049.90616.

45.

De Paepe

Lemoyne

Buylaert

. Disorders of consciousness induced by intoxication. Neurol Clin 2012; 30: 359–384. DOI: 10.1016/j.ncl.2011.10.003.

46.

Beskind

Stolz

Gross

, et al. A comparison of the prehospital motor component of the Glasgow coma scale (mGCS) to the prehospital total GCS (tGCS) as a prehospital risk adjustment measure for trauma patients. Prehosp Emerg Care 2014; 18: 68–75. DOI: 10.3109/10903127.2013.844870.

47.

Wijdicks

Kramer

Rohs

Jr. , et al. Comparison of the Full outline of UnResponsiveness score and the Glasgow coma scale in predicting mortality in critically ill patients. Crit Care Med 2015; 43: 439–444. DOI: 10.1097/ccm.0000000000000707.

48.

Okasha

Fayed

Saleh

. The FOUR score predicts mortality, endotracheal intubation and ICU length of stay after traumatic brain injury. Neurocritical Care 2014; 21: 496–504. DOI: 10.1007/s12028-014-9995-6.

49.

Zadravecz

Tien

Robertson-Dick

, et al. Comparison of mental-status scales for predicting mortality on the general wards. J Hosp Med 2015; 10: 658–663. DOI: 10.1002/jhm.2415.

50.

Kim

Lee

, et al. Clinical phenotypes of delirium in patients admitted to the cardiac intensive care unit. PLoS One 2022; 17: e0273965. DOI: 10.1371/journal.pone.0273965.

51.

Sabzghabaee

Yaraghi

Khalilidehkordi

, et al. The outcome of agitation in poisoned patients in an Iranian tertiary care university hospital. Neurol Res Int 2014; 2014: 275064. DOI: 10.1155/2014/275064.

52.

Shama

El-Gharbawy

Wahdan

, et al. Assessment of the efficacy of four scoring systems in prediction of acute organophosphorous poisoning outcome. Tanta Med J 2021; 49(3): 187–197. DOI: 10.4103/tmj.tmj_107_21.

53.

Wahdan

Helal

. Comparing the effectiveness of three scoring Systems in Outcome Prediction of acute tricyclic antidepressants poisoning cases. Ain Shams J Forensic Med Clin Toxicol 2021; 37(2): 136–149. DOI: 10.21608/AJFM.2021.185478.

54.

. Evaluation of various scoring systems in prediction of acute carbon monoxide poisoning outcome. Ain Shams J Forensic Med Clin Toxicol 2022; 38: 79–86. DOI: 10.21608/AJFM.2022.213306.

55.

Lashin

Sharif

. Evaluation of various scoring systems as predictors of the need for intensive care unit admission and other adverse outcomes among patients with acute clozapine poisoning. Toxicol Res 2023; 12: 468–479. DOI: 10.1093/toxres/tfad029.

56.

Khanal

Bhandari

Shrestha

, et al. Comparison of outcome predictions by the Glasgow coma scale and the Full Outline of UnResponsiveness score in the neurological and neurosurgical patients in the Intensive Care Unit. Indian J Crit Care Med 2016; 20(8): 473–476. DOI: 10.4103/0972-5229.188199.

57.

Bayraktar

Sahinoglu

Cicekci

, et al. Comparison of Glasgow coma scale and Full outline of unresponsiveness (four) score: a prospective study. Turk Neurosurg 2019; 29: 285–288. DOI: 10.5137/1019-5149.Jtn.24175-18.2.

58.

Healey

Osler

Rogers

, et al. Improving the Glasgow Coma Scale score: motor score alone is a better predictor. J Trauma 2003; 54: 671–678. DOI: 10.1097/01.Ta.0000058130.30490.5d.

59.

Gill

Reiley

Green

. Interrater reliability of Glasgow Coma Scale scores in the emergency department. Ann Emerg Med 2004; 43: 215–223. DOI: 10.1016/s0196-0644(03)00814-x.

60.

Brown

Forsythe

Stassen

, et al. Evidence-based improvement of the national trauma triage protocol: the Glasgow coma scale versus Glasgow coma scale motor subscale. J Trauma Acute Care Surg 2014; 77: 95–102. DOI: 10.1097/ta.0000000000000280.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.09 MB