a b s t r a c t

Background: Although many studies have evaluated the diagnostic value of the National Emergency X-ray Utilization Studies (NEXUS) chest rules in assessment of traumatic chest injuries, there still is no consensus on this subject matter. Therefore, this systematic review and meta-analysis aims to review the current existing lit- erature in order to evaluate the diagnostic value of NEXUS chest rules for assessment of traumatic chest injuries. Method: Databases of Medline, Embase, Scopus and Web of Science were searched until August 20th, 2022. Two independent reviewers screened the articles related to the diagnostic value of NEXUS chest radiography, NEXUS Chest CT-all and NEXUS chest-Major.

Results: Data of 6 studies, on 23,741 patients, were included in this review. Since only one article assessed the value of NEXUS Chest CT scan, the meta-analysis was performed only on NEXUS chest radiography rule. Pooled analysis on the results of 5 articles showed that the AUC of NEXUS chest radiography rule in assessment of trau- matic chest injuries was 0.98 (95% CI: 0.96 to 0.99), with a sensitivity and specificity of 0.99 (95% CI: 0.98 to 0.99) and 0.32 (95% CI: 0.17 to 0.52), respectively. Positive and negative likelihood ratio of NEXUS chest radiography rule were 1.46 (95% CI: 1.12 to 1.90) and 0.04 (95% CI: 0.03 to 0.06). Overall Diagnostic odds ratio was calculated to be 36.67 (95% CI: 19.17 to 70.16).

Conclusion: Our findings indicate that NEXUS chest radiography rule is a sensitive decision rule for assessment of traumatic chest injuries, but its specificity was found to be low. However, few articles have investigated the diagnostic value of NEXUS chest rules, especially the NEXUS chest CT scan, and more studies need to be done in order to strengthen the currently provided results.

(C) 2023

1. Introduction

Trauma is the physical force inflicted when a person is exposed to sudden or acute degrees of energy [1]. Trauma is one of the most important causes of death in people younger than 50 years [2-4] and also a cause of debilitating Functional limitations [5,6]. Traumatic chest injuries are the cause of up to 50% of deaths due to trauma [7].

* Correspondence to: M. Yousefifard, Physiology Research Center, Iran University of Medical Sciences, Hemmat Highway, P.O Box: 14665-354, Tehran, Iran.

?? Correspondence to: S. Safari, Emergency Department, Shohadaye Tajrish Hospital, Tajrish Square, Tehran, Iran.

E-mail addresses: yousefifard.m@iums.ac.ir (M. Yousefifard), s.safari@sbmu.ac.ir (S. Safari).

Hemopneumothorax, hemothorax and rib fracture are the most important cause of death following traumatic thoracic injuries [8].

According to the current guidelines chest radiographic study of each multiple trauma patient is a must. However, in most Multiple trauma patients these studies yield no significant abnormalities. Safari et al. [9], reported that 9.2% of the chest radiography studies on multiple trauma patients had significant findings. In a similar study Yousefifard et al. [10], reported a rate of 12.6%. Therefore, it seems that >80% of the chest radiographies done on multiple trauma patients are unnecessary.

These redundant studies lead to non-essential exposure to x-ray and also overcrowding of emergency departments. The high number of radiographies performed in the emergency wards also causes a longer patient management time. Having no portable x-ray machine and the inability of transferring critically ill patients to the radiology department, are other problems faced with depending on a frequently

https://doi.org/10.1016/j.ajem.2022.12.038

0735-6757/(C) 2023

Table 1

Definition of national emergency X-ray utilization studies (NEXUS) chest rules.

Variable	NEXUS chest radiography	NEXUS chest CT-all	NEXUS chest CT-major
Age > 60 years	?	–	–
Rapid deceleration mechanism	?	?	–
Chest pain	?	–	–
Intoxication	?	–	–
Altered mental status	?	–	–
Distracting painful injury	?	?	?
Chest wall tenderness	?	?	?
Abnormal chest radiography	–	?	?
Sternum tenderness	–	?	?
thoracic spine tenderness	–	?	?
Scapula tenderness	–	?	?

unnecessary radiography for management of traumatic patients. These unnecessary radiographies have an immense amount of Financial burden on the healthcare system.

In order to reduce the number of radiographies requested in the emergency units a few tools and scoring systems have been proposed; some of which are National Emergency X-ray Utilization Studies (NEXUS) rules and thoracic injury Rule out Criteria (TIRC) [9,11-13]. NEXUS chest rules include NEXUS chest radiography, NEXUS chest CT-All and NEXUS chest CT-Major. These criteria are demonstrated with each of their components in Table 1. Studies performed on using these scoring systems demonstrate a significant decrease in the amount of chest radiographies done in multiple trauma patients [9]. Although many studies have been done to evaluate the diagnostic value of the NEXUS chest rules in assessment of traumatic chest injuries, there still is no comprehensive review on this subject matter. Therefore, this sys- tematic review and meta-analysis aims to review the current existing literature in order to evaluate the diagnostic value of NEXUS chest rules for assessment of traumatic chest injuries.

1. Method
   1. Study design

This systematic review and meta-analysis aimed to investigate the relevant original studies performed on the evaluation of the diagnostic value of NEXUS chest rules in assessment of traumatic chest injuries. PICO was defined as follows: Population (P): Diagnostic accuracy stud- ies performed on NEXUS chest rules in multiple trauma patients. Index test (I): NEXUS chest rules. Comparison (C): findings of chest radiogra- phies. Outcome (O): assesment of chest traumatic injuries.

• 1. Search strategy

Keywords related to NEXUS chest rules and traumatic chest injuries were chosen based on the MeSH and Emtree of Medline and Embase databases, consulting expert opinion and reviewing the existing litera- ture. Four databases of Medline, Embase, Scopus and Web of Science were searched until August 20th, 2022. Search queries used in this study are provided in Supplementary material. Google and Google Scholar search engines and citations of related articles were manually search in order to further identify non-indexed papers.

• 1. Selection criteria

Diagnostic accuracy studies, prospective studies and adult popula- tion were the inclusion criteria of this review. We included all studies on the diagnostic value of NEXUS chest radiography, NEXUS chest CT-All and NEXUS chest CT-Major for assessment of traumatic chest injuries.

The exclusion criteria were radiographic imaging performed before evaluation of patient with NEXUS chest rules, review articles, duplicate

Fig. 1. PRISMA flow diagram of current meta-analysis.

Table 2

Characteristics of included papers.

Study

Design

Sample size

Age?

male

assessor

injury to test interval (hrs)

Injured / normal (n)

NEXUS rule

TP

FP

FN

TN

Acar, 2020, Turkey [17] Norouzi, 2019,

Prospective, Cohort Prospective,

690 1925

45.8 +- 16.9 43.7 +- 9.16

406 1059

EP EP

<24 <24

178 / 512 362 / 1563

NEXUS radiography rule NEXUS

1780 910

316 883

30 20

193 949

Iran [16] Rodriguez. 2013,

Cross-sectional Prospective, Cohort

9905

46

6220

EP

<24

1478 / 8427

radiography rule NEXUS

1461

7304

17

1123

USA [11] Rodriguez, 2011,

Prospective, Cohort

2628

45.0 +- 19.8

1698

EP

<24

271 / 2357

radiography rule NEXUS

2690

2028

20

329

USA [12] Safari, 2018, Iran

Prospective,

3118

37.4 +- 16.9

1790

EP

<24

287 / 2831

radiography rule NEXUS

2830

1134

40

1698

[9] Rodriguez, 2015,

Cross-sectional Prospective, Cohort

2628

45 (28-61)

NR

EP

<24

701 / 1927

radiography rule NEXUS CT-All

669

1436

32

491

USA [13]

NEXUS CT-Major

CT: Computed tomography; EP: Emergency medicine physician; FN: False negative; FP: False positive; NEXUS: The National Emergency X-Radiography Utilization Study; TN: True negative; TP: True positive.

* Age was reported as mean +- SD or median (IQR).

Table 3

Risk of bias and applicability of included studies.

Author	Risk of bias					Applicability			Overall
	Patients’ selection	Index test	Reference test	Flow and timing		Patients’ selection	Index test	Reference test
Acar, 2020	High	Low	Low	Unclear		Low	Low	Low	Some concern in risk of bias
Norouzi, 2019	High	Low	Low	Unclear		Low	Low	Low	Some concern in risk of bias
Rodriguez, 2013	High	Low	Low	Unclear		Low	Low	Low	Some concern in risk of bias
Rodriguez, 2011	High	Low	Low	Unclear		Low	Low	Low	Some concern in risk of bias
Sadari, 2018	High	Low	Low	Unclear		Low	Low	Low	Some concern in risk of bias
Rodriguez, 2015	High	Low	Low	Unclear		Low	Low	Low	Some concern in risk of bias

records, not investigating the diagnostic value of NEXUS chest rules and not investigating traumatic chest injuries.

• 1. Data collection

After removal of duplicates from the records retrieved from the databases, two independent reviewers screened the titles and abstracts

Fig. 2. Summary receiver operating characteristic of NEXUS chest radiography in diagnosis of thoracic injuries.

and included the relevant articles. Full texts of these articles were studied, and most relevant articles were selected based on the selection criteria.

Data provided by the included articles were extracted into a pre-designed checklist based on PRISMA guidelines [14].The extracted data were first author name, Publication year, sample size, number of patients with and without traumatic chest injuries, the person reporting the radiography or CT scan, the person evaluating NEXUS chest rule in the patients and true positive, true negative, false positive and false negative values reported by the articles.

• 1. Quality assessment

Quality of the included articles were evaluated using QUADAS-2 guidelines [15]. Interrater reliability was investigated in order to assess the agreement between the two reviewers. Any disagreements were resolved by consulting a third reviewer.

• 1. Statistical analysis

Analysis was performed using STATA 14.0 statistical analysis software. Data were input as true positive, false positive, true negative and false negative. Receiver operating characteristic (ROC) curve, specificity, sensitivity, positive and negative likelihood ratio, odds ratio was calculated using the “midas” command. The heterogeneity between the included articles was evaluated by chi squared and I² statistics and Deek’s asymmetry plot test was used to assess the publication bias.

1. Results
   1. Study characteristics

Our search query resulted in 522 records. From which 14 articles were included in this study After removal of duplicates and title and

Fig. 3. Sensitivity and specificity of NEXUS chest radiography in diagnosis of thoracic injuries.

abstract screening, 14 articles were judged to be applicable. Three addi- tional studies were found by the manual search, making a total of 17 included articles. 6 of these articles were finally chosen to be included in the meta-analysis [9,11-13,16,17]. Eleven articles were excluded due to the following reasons: Studies of the same registry in five (three studies in systematic search and two in manual search), not

assessing NEXUS chest rules in four, not being a diagnostic accuracy study in one and assessment of aortic injury only in one articles (Fig. 1). Five articles were performed on NEXUS chest radiography rule and one study had used NEXUS chest CT as the assessment tool. Three arti- cles were done in united states of America, two in Iran and one in Turkey. Two of the studies were cross-sectional, with the remaining

Fig. 4. Positive and negative diagnostic likelihood ratio (DLR) of NEXUS chest radiography in diagnosis of thoracic injuries.

Fig. 5. Diagnostic odds ratio of NEXUS chest radiography in diagnosis of thoracic injuries.

four designed as cohort studies. All the studies were performed on blunt chest trauma patients.

The included studies had data on 20,894 patients, 61% of them being male. The average age of each of the included populations was between

37.4 and 46 years. Nexus criteria and radiographic assessments were performed by emergency physician and in the first 24 h after trauma in all studies. Table 2 provides more detailed information on the charac- teristics of each of the included studies. Since only one articles assessed the value of NEXUS chest CT scan, the meta-analysis was performed only on the five articles NEXUS chest radiography rule.

• 1. Bias assessment“>Risk of bias assessment

QUADAS-2 risk of bias assessment guideline was used to assess the quality of the included articles. Patient selection was rated to be high risk of bias in all the studies, due to the convenience sampling method used. The time period between NEXUS criteria assessment and

Fig. 6. Publication bias among included studies.

radiographic studies were not reported in any of the studies, and thus domain of flow and timing was rated as unclear. All the studies were low risk in domain of applicability. Overall, the included studies were evaluated to have low risk of bias in applicability domains and some concern in risk of bias domains of the guideline (Table 3).

• 1. Diagnostic value of NEXUS chest radiography rule in assessment of traumatic chest injuries

Our pooled analysis demonstrated that AUC of NEXUS chest radiography rule in assessment of traumatic chest injuries was 0.98 (95% CI: 0.96 to 0.99) (Fig. 2), with a sensitivity and specificity of 0.99

(95% CI: 0.98 to 0.99) and 0.32 (95% CI: 0.17 to 0.52), respectively

(Fig. 3). Positive and negative likelihood ratio of NEXUS chest radiogra- phy rule were 1.46 (95% CI: 1.12 to 1.90) and 0.04 (95% CI: 0.03 to 0.06) (Fig. 4). Overall diagnostic odds ratio was calculated to be 36.67 (95% CI: 19.17 to 70.16) (Fig. 5).

• 1. Publication bias

No publication bias was observed between the included articles (p = 0.992) and Deek’s funnel plot demonstrated that the studies included in the meta-analysis had near symmetric distribution (Fig. 6).

1. Discussion

The results of this study show that NEXUS chest radiography rule is a sensitive tool for assessment of traumatic chest injuries but its specific- ity was found to be low. However, only five studies have investigated the diagnostic value of NEXUS chest rules and more studies need to be done in order to strengthen the currently provided results.

An interesting finding in the current study is the diverse results for NEXUS chest radiography rule in different populations. Studies done on the evaluation of this tool in American populations report a specific- ity of 0.13 and 0.14, while studies done in Iranian populations report specificities of 0.52 and 0.60. Another study done in Turkey demon- strated a specificity of 0.38 for NEXUS chest radiography. This diversity which can be due to reasons such as population differences and the

possible difference in the expertise of the physicians in charge, show the need for more studies before clinical implementation of this tool.

A limitation of NEXUS chest radiography rule is taking the vehicle speed and fall height into consideration. The NEXUS chest radiography may not be applicable in low and middle -income countries, since such information (vehicle speed and fall height) is not accounted for in Developing countries. Another limitation of the rule is including the distracting injury. Pain is a subjective variable and its presentation and expression is affected by cultural and personal diversities. In addition, concept of distracting pain may be confusing since it is not precisely defined in the NEXUS chest radiography derivation study.

Based on our knowledge, only one study has been performed on

NEXUS chest CT, which is the validation and derivation study performed by Rodriguez et al. in 2016. Two decision making tools of NEXUS chest CT-All and NEXUS chest CT-Major were derivate and validated in this study. The difference in these two tools is that the rapid deceleration mechanism information is not necessary in evaluation of NEXUS CT-Major. Results of the validation section of this study indicated that sensitivity of NEXUS chest CT-All and CT-Major in assessment of trau- matic chest injuries were both 95.4 and 90.7% and their specificities were 25.5 and 37.9% respectively [13]. In the validation study done by Rodriguez et al., data for 2628 patients of the original 5475 included pa- tients was analyzed due to the lack of both radiographic and CT imaging data on all patients. This shows that the participation rate of the study was roughly 50%. With this serious limitation and with only one article evaluating the use of NEXUS chest CT, this tool has been introduced as a screening tool in the Rosen’s Emergency Medicine: Concepts and Clinical Practice; a textbook in the emergency medicine [18]. Consider- ing the low level of current evidence on the efficacy of NEXUS chest CT scan for assessment of traumatic chest injuries, we suggest that this issue be addressed in the textbook until further investigation is made on the diagnostic value of NEXUS chest CT in other populations.

The scarce number of included articles was a limitation of this study. 17 articles were found to be potentially eligible, but many of the articles were extracted from NEXUS database and five of the articles were excluded due to being duplicates of NEXUS registry. So far only six stud- ies have evaluated the efficacy of the NEXUS chest radiography tool, whereas more studies are needed for the validation of a new tool. From the six included studies, three were performed on American pop- ulation, two on Iranian population and one in Turkish population and these articles themselves have conflicting results on the efficacy of the NEXUS chest radiography tool.

1. Conclusion

Our results indicate that NEXUS chest criteria is a sensitive tool for assessment of traumatic chest injury, although it has low reported rates for specificity. So far only six articles have investigated the NEXUS chest criteria with only one of the studies evaluating the NEXUS chest CT criteria and so there is a need for further investigation on the efficacy of NEXUS chest criteria and other similar decision- making rules in assessment of traumatic chest injuries.

Fund

This study was funded and supported by Research Center for Trauma in Police Operations, Directorate of Health, Rescue & Treatment, Police Headquarter, Tehran, Iran.

Authors’ contribution

Study design: Mahmoud Yousefifard, Saeed Safari.

Data gathering: Koohyar Ahmadzadeh, Mohammad Abbasi, Mahmoud Yousefifard.

Analysis: Mahmoud Yousefifard. Drafting: Koohyar Ahmadzadeh.

Critically revised: All authors.

CRediT authorship contribution statement

Koohyar Ahmadzadeh: Writing – original draft, Validation, Investigation. Mohammad Abbasi: Writing – original draft, Investiga- tion. Mahmoud Yousefifard: Writing – review & editing, Validation, Supervision, Project administration, Methodology, Investigation, Formal analysis, Conceptualization. Saeed Safari: Writing – review & editing, Supervision, Investigation, Conceptualization.

Declaration of Competing Interest

The authors declared that there is no conflict of interest.

Acknowledgment

None.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2022.12.038.

References

1. Subcommittee ATLS, and International ATLS Working Group. Advanced trauma life support (ATLS(R)). J Trauma Acute Care Surg. 2013;74(5):1363-6.
2. World Health Organization. World Health Organization issues injury surveillance guidelines. N S W Public Health Bull. 2002;13(4):83.
3. Saadat S, Yousefifard M, Asady H, Jafari AM, Fayaz M, Hosseini M. The most impor- tant causes of death in Iranian population; a retrospective cohort study. Emergency. 2015;3(1):16-21.
4. Dyer O. One million people die on world’s roads every year. BMJ. 2004;328(7444): 851.
5. Sauaia A, Moore FA, Moore EE, Moser KS, Brennan R, Read RA, et al. Epidemiology of Trauma deaths. J Trauma Inj Infect Critic Care. 1995;38(2):185-93.
6. Knudsen AK, Allebeck P, Tollanes MC, Skogen JC, Iburg KM, McGrath JJ, et al. Life ex- pectancy and Disease burden in the Nordic countries: results from the global burden of diseases, injuries, and risk factors study 2017. Lancet Public Health. 2019;4(12): e658-69.
7. Langdorf MI, Medak AJ, Hendey GW, Nishijima DK, Mower WR, Raja AS, et al. Prevalence and clinical import of thoracic injury identified by chest computed tomography but not chest radiography in blunt trauma: multicenter prospective co- hort study. Ann Emerg Med. 2015;66(6):589-600.
8. Yimam AE, Mustofa SY, Aytolign HA. Mortality rate and factors associated with death in traumatic chest injury patients: a retrospective study. Int J Surg Open. 2021; 37:100420.
9. Safari S, Radfar F, Baratloo A. Thoracic injury rule out criteria and NEXUS chest in predicting the risk of traumatic intra-thoracic injuries: a diagnostic accuracy study. Injury. 2018;49(5):959-62.
10. Yousefifard M, Hosseini M, Parvizi MR. Pediatric thoracic injury rule out criteria (pTIRC) in diagnosis of Very low risk children for traumatic intrathoracic injuries; a diagnostic accuracy study. Archiv Academ Emerg Med. 2020;8(1):e7.
11. Rodriguez RM, Anglin D, Langdorf MI, Baumann BM, Hendey GW, Bradley RN, et al. NEXUS chest: validation of a decision instrument for selective Chest imaging in blunt trauma. JAMA Surg. 2013;148(10):940-6.
12. Rodriguez RM, Hendey GW, Mower W, Kea B, Fortman J, Merchant G, et al. Deriva- tion of a decision instrument for selective chest radiography in blunt trauma. J Trauma: Inj Infect Critic Care. 2011;71(3):549-53.
13. Rodriguez RM, Langdorf MI, Nishijima D, Baumann BM, Hendey GW, Medak AJ, et al. Derivation and validation of two Decision instruments for selective chest CT in blunt trauma: a Multicenter prospective observational study (NEXUS chest CT). PLoS Med. 2015;12(10):e1001883.
14. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for sys- tematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6(7): e1000097.
15. Whiting PF. QUADAS-2: a revised tool for the quality assessment of diagnostic accu- racy studies. Ann Intern Med. 2011;155(8):529-36.
16. Norouzi N, Amini A, Hatamabadi H. Comparison of diagnostic accuracy of NEXUS chest and thoracic injury rule-out criteria in patients with blunt trauma. A Cross- Sect Study Trauma Monthly. 2019;24(3):1-6.
17. Acar E. Evaluation of Nexus X-ray rules in blunt thorax trauma. Turk J Trauma Emerg

Surg. 2020;26(6):920-6.

1. Marx JM, Hockberger RS, Walls RM. Rosen’s emergency medicine: concepts and clin- ical practice. Fifth Ed Ann Emerg Med. 2003;41(5):769-70.