Identification of a reliable sacral-sparing examination to assess the ASIA impairment scale in patients with traumatic spinal cord injury

Abstract

Objectives

We evaluated the time course of the American Spinal Cord Injury Association (ASIA) impairment scale (AIS) for up to three months in participants within 72 h after traumatic spinal cord injury (TSCI) with complete paralysis. We aimed to determine the most useful sacral-sparing examination (deep anal pressure [DAP], voluntary anal contraction [VAC], S4-5 light touch [LT], or pin prick [PP] sensation) in determining AIS grades.

Design

Retrospective cohort study.

Setting

Spinal Injuries Center, Fukuoka, Japan.

Participants

Among 668 TSCI participants registered in the Japan Single Center study for Spinal Cord Injury Database (JSSCI-DB) between January 2012 and May 2020, we extracted the data of 80 patients with AIS grade A within 72 h after injury and neurological level of injury (NLI) at T12 or higher.

Interventions

None.

Outcome measures

The sacral-sparing examination at the time of the change to incomplete paralysis was compared to the AIS determination using a standard algorithm and with each assessment including the VAC, DAP, S4-5LT, and S4-5PP examinations at the time of AIS functional change. Agreement among assessments was evaluated using weighted kappa coefficients. The relationship was evaluated using Spearman’s rank correlation coefficients.

Results

Fifteen participants (18.8%) improved to incomplete paralysis (AIS B to D) within three months after injury. The single assessment among the sacral-sparing examinations with the highest agreement and strongest correlation with AIS determination was the S4-5LT examination (k = 0.89, P < 0.01, r = 0.84, P < 0.01).

Conclusions

The S4-5LT examination is key in determining complete or incomplete paralysis due to its high discriminatory power.

Introduction

The International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) worksheet developed by the American Spinal Injury Association (ASIA) and the International Society for Spinal Cord Injury (ISCoS) is the major standard assessment for classifying neurological function in traumatic spinal cord injury (TSCI) and is widely used globally.^1,2 The ASIA Impairment Scale (AIS) in the ISNCSCI worksheet classifies functional impairment on a scale from A to E according to the degree of spinal cord injury. To determine complete paralysis (AIS grade A), incomplete paralysis (AIS grade B–D), or normal (AIS grade E), evaluation of the following tests is essential: sacral-sparing with voluntary anal contraction (VAC), deep anal pressure (DAP), S4-5 light touch sensation (LT), and pin prick sensation (PP). In addition, if the individual has an incomplete sensory classification, the presence of VAC or motor function more than three levels below the normal motor level on a given side is an ancillary condition in determining AIS grades B, C, and D.³ Neurological function recovers in some individuals over time after TSCI, and most of the recovery occurs in the acute phase.^4–6 Kawano et al. studied the recovery of complete cervical spinal injuries and observed that, when AIS changed from complete to incomplete, the change occurred during the first 8 weeks after injury.⁶ However, few detailed studies have evaluated the sacral-sparing examination, which is the most important factor to determine AIS in the process of the change in functional status from complete to incomplete paralysis after TSCI in the acute phase. Evaluation of the sacral-sparing examination at the time of AIS determination in the chronic phase has been reported.⁷ However, the details and reliability of the sacral-sparing examination focusing on the acute phase in which neurological function recovery is more active have not been shown. In addition, DAP, one of the sacral-sparing examinations in the ISNCSCI worksheet, is reportedly limited by the inconsistent interpretation of terms and a lack of standardization of methods among examiners.⁸ Moreover, the reliability and validity of anal sensation have not been well studied.⁸ Strong stimulation of the anorectal wall may result in a vague sensation that is relayed via the autonomic nervous system rather than via the somatosensory system.⁹ Few studies have examined the evaluation items among the sacral-sparing examinations to assess their usefulness in determining AIS.^7,10 Therefore, our hypothesis was that there might be items with high discriminatory performance that can be used to determine AIS grades without having to administer all four sacral-sparing examinations. In previous studies, LT has been used as a predictor variable for the prognosis of gait and bladder function after TSCI and was found to be a highly reliable evaluation methodology.^11–14 Accordingly, we hypothesized that S4-5LT is the most appropriate method for determining the AIS grade in the evaluation of the sacral-sparing examinations. The purpose of this study was to investigate the process of recovery from complete paralysis in acute TSCI participants over time and to clarify the degree of agreement among the AIS determinations using the standard ISNCSCI worksheet algorithm (Version 2 Beta) and the AIS determination using each single evaluation of the sacral-sparing examination at the time of AIS functional change.

Methods

Study design and participants

This retrospective cohort study was performed at the Spinal Injuries Center at Fukuoka in Japan. Participants with TSCI who were hospitalized between January 2012 and May 2020 and were registered in the Japan Single Center study for Spinal Cord Injury Database (JSSCI-DB) were included in this study. Participants underwent surgery if required, and rehabilitation treatment was continued until discharge. The JSSCI-DB was designed to prospectively collect data on the neurological evaluation of patients with TSCI over time, and the inclusion and exclusion criteria of this database were described in detail.^15,16 All data were managed by ASIA members who had completed the International Standards Training E Program (InSTeP). We extracted data from the JSSCI-DB and conducted a retrospective analysis.

Among 668 participants registered in the JSSCI-DB, 80 participants with TSCI, including those with an AIS grade of A, those with a neurological level of injury (NLI) of T12 or higher within 72 h after injury, and those who could be followed for at least three months after injury, were selected as participants in this study. Based on the possibility that people with TSCI who were classified as AIS grade A may continue to recover from AIS grade A within the first two months after injury,⁶ the follow-up period of those participants would be slightly extended from the range of two months, and they would receive up to three months of follow-up after the injury. Patients with an AIS grade of B–E, those with an NLI L1 of S4, 5 within 72 h after injury, and those who could not be followed up for at least three months after injury were excluded. A detailed flowchart of the participants who met the inclusion criteria for this study is shown in Fig. 1. We certify that all applicable institutional and governmental regulations concerning the ethical use of human volunteers were followed during this research. Informed consent was provided by all participants, and this study was approved by the Institutional Ethical Review Board (approval number: 15-2). In addition, the sample size, which included the number of participants who met the inclusion criteria of this study, was reasonable to derive an appropriate effect size for statistical analysis.

Figure 1.

Flowchart of the selection of participants for analysis.

Note: From the 668 participants who were registered in the Japan Single Center study for Spinal Cord Injury Database (JSSCI-DB) between January 2012 and May 2020, we excluded 186 who were not hospitalized within 72 h after injury. From the remaining 482 participants, we excluded 346 who had incomplete paralysis (ASIA impairment scale grade B to E) and four with a neurological level of injury (NLI) from L1 to S4, 5 at 72 h after injury. In addition, 52 participants who could not be followed up for three months after injury were excluded. Finally, 80 participants who met the inclusion criteria were included in this study.

Evaluations

The basic medical and neurological details of the selected participants (such as key muscles, LT, PP, VAC, DAP, AIS, and NLI) in the longitudinal ISNCSCI worksheets at 72 h; 2, 4, and 6 weeks; and 2 and 3 months after injury were extracted from the JSSCI-DB. In addition, we collected data on the change in functional status from complete to incomplete paralysis, focusing on the evaluation of the sacral-sparing examinations. The test error range of the longitudinal evaluation in this study was ± 12 h (72 h) and ±7 days (2 weeks, 4 weeks, 6 weeks, 2 months, and 3 months). All the information provided in the longitudinal ISNCSCI worksheet was evaluated and described according to the ISNCSCI guidelines.³

AIS determination method employed for verification

After assessing the longitudinal data in the ISNCSCI worksheet, we investigated the details of the sacral-sparing examinations and the time period in which the participant’s functional status had changed from complete to incomplete paralysis. In the ISNCSCI algorithm, the results of the sensory tests (LT and PP) of the S4-5, VAC, and DAP examinations were used to determine complete or incomplete paralysis. In this verification, AIS assignment was performed using each assessment of the four sacral-sparing examinations, which was then compared with the AIS assignment using the standard ISNCSCI algorithm. The flowchart of the AIS determination method for the verification of the AIS assignment is shown in Fig. 2.

Figure 2.

Flowchart illustrating the method for determining the American Spinal Cord Association impairment scale for verification.

Note: In the standard ISNCSCI algorithm, the American Spinal Cord Association impairment scale (AIS) is determined by performing all four sacral-sparing examinations. The AIS is determined according to the flowchart.

Statistical analyses

Based on the information in the ISNCSCI worksheet at the time of the AIS functional change, the degree of agreement between the AIS assignment using the standard algorithm and the AIS assignment using each assessment of the four sacral-sparing examinations was evaluated using Cohen’s weighted kappa coefficient (k). The relationship between the evaluated items was analyzed using Spearman’s rank correlation coefficient (r). The Landis and Koch criteria were used to interpret the kappa coefficient.¹⁷ The interpretations were as follows: 0, poor; 0.01–0.20, slight; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, substantial; and 0.81–1.00, almost perfect. In addition, the receiver operating characteristic (ROC) curve was obtained to evaluate the ROC curvilinear area, (area under the curve [AUC]). The AUC was defined as whether each single sacral-sparing examination could determine complete paralysis (AIS grade A) or incomplete paralysis (AIS grades B–D). Statistical analyses were performed using MATLAB R2020aTM (The MathWorks Inc., Natick, MA, USA) and R version 4.3.3 (The R Foundation, Vienna, Austria). A P-value < 0.05 was considered statistically significant.

Results

Of the 668 participants who were registered in the JSSCI-DB between January 2012 and May 2020, 80 met the inclusion criteria for this study, and their data were selected for analysis. The details of participant eligibility are shown in Fig. 1. In addition, details of the characteristics of the participants are shown in Table 1 (such as age at injury, sex, cause of injury, AIS and NLI at 72 h and three months after injury, and the time point of improvement in participants with recovery from complete paralysis).

Table 1.

Characteristics of the participants.

Characteristics	n (percentages)
Number of participants	80
Age at injury	56.3 ± 17.1(18-80)
Sex (Male / Female)	68(85) / 12(15)
Traumatic etiology
Falling from a height	43(53.8)
Vehicular accidents	20(25)
Fall down	8(10)
Sports	3(3.8)
Other causes of trauma	6(7.5)
ASIA impairment scale at 72 h after injury
Grade A	80(100)
ASIA impairment scale at 3 months after injury
Grade A	65(81.3)
Grade B	9(11.3)
Grade C	4(5)
Grade D	2(2.5)
Time of the change to incomplete paralysis
2 weeks	6(7.5)
4 weeks	6(7.5)
6 weeks	1(1.3)
2 months	1(1.3)
3 months	1(1.3)
Neurological level of injury at 72 h after injury
C1–C8	60(75)
T1–T12	20(25)
Neurological level of injury at 3 months after injury
C1–C8	61(76.3)
T1–T12	16(20)
L1–S5	3(3.8)

Note: Age at injury is expressed mean ± standard deviation (ranges). Other values were expressed as n, (percentages).

After observing the change in function from 72 h to three months after the injury, 15 participants (18.8%) recovered from complete paralysis. The time period of the change in function to recovery from complete paralysis, along with the NLI, the details of the evaluation of the sacral-sparing examinations, and the changes in the AIS over time are shown in Table 2. Among those who recovered from complete paralysis, 12 participants (80%) had early recovery to incomplete paralysis within four weeks post-injury. The percentages of each sacral-sparing examination that were assessed at the time point of the change to incomplete paralysis within three months post-injury are shown in Fig. 3. Of the participants with a change to incomplete paralysis, 11 of 15 (73.3%) participants had preserved an S4-5LT, and 10 of 15 (66.7%) had preserved a DAP.

Table 2.

Details of changes in the participants’ sacral-sparing examinations and American Spinal Cord Association impairment scale (AIS) over time with a functional change to incomplete paralysis.

Participants	NLI (At the point of change to incomplete paralysis)	The time point of changed to incomplete paralysis	Details of sacral-sparing examinations during the change to incomplete paralysis				Changed in the ASIA impairment scale over time
			DAP	VAC	S4-5LT sensation	S4-5PP sensation	72 h	2 weeks	4 weeks	6 weeks	2 months	3 months
1	C2	2 weeks	Y	N	N	N	A	B	B	B	B	B
2	C4	2 weeks	Y	N	N	N	A	B	B	B	B	B
3	T12	2 weeks	Y	N	Y	N	A	B	B	B	B	B
4	C3	2 weeks	Y	N	Y	N	A	B	B	B	B	B
5	C4	2 weeks	Y	N	Y	Y	A	B	B	B	B	B
6	C4	2 weeks	N	N	Y	N	A	D	D	D	D	D
7	C5	4 weeks	Y	N	Y	N	A	A	B	B	B	B
8	C8	4 weeks	N	N	Y	N	A	A	B	B	B	B
9	C5	4 weeks	Y	N	Y	N	A	A	B	B	B	B
10	C6	4 weeks	N	N	Y	N	A	A	B	C	C	C
11	C2	4 weeks	Y	N	N	N	A	A	C	C	C	C
12	C4	4 weeks	Y	Y	Y	Y	A	A	C	C	C	C
13	T12	6 weeks	Y	N	N	N	A	A	A	B	B	B
14	C6	2 months	N	N	Y	N	A	A	A	A	C	C
15	C4	3 months	N	N	Y	N	A	A	A	A	A	D

Abbreviations: NLI, neurological level of injury; DAP, deep anal pressure; VAC, voluntary anal contraction; LT, light touch sensation; PP, pin prick sensation; N, no; Y, yes.

Figure 3.

Percentage of each sacral-sparing examination at the time point of recovery from complete paralysis by three months post-injury.

Note: The detailed sacral-sparing examinations at the time point of recovery from complete paralysis show a 73.3% improvement in the S4-5 LT, followed by a 66.7% recovery in the DAP. Abbreviations: DAP, deep anal pressure; VAC, voluntary anal contraction; LT, light touch sensation; PP, pin prick sensation.

Table 3 shows the agreement and correlation between the results of the AIS determinations using the standard ISNCSCI algorithm and the results of the AIS determination using only one of the assessments of the four sacral-sparing examinations. Of the four sacral-sparing examinations, S4-5LT had a k = 0.89 (P < 0.01), which was interpreted as almost perfect with the AIS determination of the standard ISNCSCI algorithm. The correlation between the AIS determination of the standard ISNCSCI algorithm and the AIS determination using only the S4-5LT as the assessment of the sacral-sparing examination was strong (r = 0.84, P < 0.01).

Table 3.

The agreement and correlation between the standard algorithm and each single assessment of the four sacral-sparing examinations.

ISNCSCI algorithm vs. Single assessment items	Weighted kappa’s coefficients		Spearman’s rank correlation coefficients
	k	95% CI	r	P-value
DAP	0.51	−0.06–1.09	0.76	P < 0.01
VAC	0.09	−0.78–0.95	0.26	P < 0.05
S4-5 LT	0.89	0.65–1.13	0.84	P < 0.01
S4-5 PP	0.19	−0.06–1.09	0.34	P < 0.01

The classification accuracy in determining complete paralysis (AIS grade A) or incomplete paralysis (AIS grade B–D) with each of the assessments of the sacral-sparing examinations using ROC analysis is shown in Table 4. The most accurate distinction was the S4-5LT (AUC: 0.87, 95% confidence interval, 0.75–0.98), followed by the DAP (AUC: 0.83, 95% confidence interval, 0.71–0.96).

Table 4.

The classification accuracy in determining complete or incomplete paralysis using each sacral-sparing examination.

Sacral-spring examination assessment items	area under the curve	95% confidence interval
		Lower bound	Upper bound
DAP	0.83	0.71	0.96
VAC	0.53	0.47	0.6
S4-5 LT	0.87	0.75	0.98
S4-5 PP	0.57	0.48	0.66

Abbreviations: DAP, deep anal pressure; VAC, voluntary anal contraction; LT, light touch sensation; PP, pin prick sensation.

Discussion

We investigated the changes in the AIS for up to three months after injury in participants with TSCI with complete paralysis during the acute phase and determined one of the most reliable sacral-sparing examinations for the determination of AIS. In this study, 18.8% of participants with an AIS of grade A at 72 h post-injury experienced a functional change to incomplete paralysis (AIS grades B–D) by three months post-injury. This result is similar to the results of previous studies conducted by Khorasanizadeh et al. and Aimetti et al.^18,19 In addition, our study found that 80% of the individuals with an improvement to incomplete paralysis recovered within four weeks after injury, and their S4-5LT and DAP evaluations improved from the early phase. However, as shown in Fig. 3, few of our participants demonstrated improvements in the VAC and S4-5PP, which are considered to be less sensitive as measurements for AIS determination. The rationale for this was that the discriminant accuracy of VAC and S4-5PP was statistically unreliable for determining complete or incomplete paralysis, as shown in Table 4. In contrast, S4-5LT and DAP had a superior discriminative performance for the AIS determination. In contrast, each examiner applies different pressure at slightly different points on the rectal and anal walls for DAP examination.⁸ Therefore, the S4-5LT examination may be a more standardized, simple, and non-invasive method of evaluation. In addition, Chun et al. reported substantial agreement between the questionnaire-based S4-5LT examination and the standard ISNCSCI method for a sacral-sparing examination, suggesting that the S4-5LT examination is a clinically manageable method.²⁰

In Table 3, the kappa coefficient of the S4-5LT was almost perfect (k = 0.89) with the standard ISNCSCI algorithm for AIS determination, along with a strong correlation (r = 0.84). These results suggest that the S4-5LT examination was the most reliable assessment among the sacral-sparing examinations that could produce results similar to those of the standard ISNCSCI algorithm for determining AIS. These findings lend support to our hypotheses.

Regarding the AIS determination using the standard ISNCSCI algorithm, all four sacral-sparing examinations are recommended. It is necessary to perform all four sacral-sparing examinations if the VAC is not testable and if there are no motor functions of more than three levels below the normal motor level on a given side, because the AIS will not be determinable and will be difficult to classify. The disadvantage of this condition is that the actual AIS grade is not determinable until all the items in the ISNCSCI worksheet have been examined. In our verification results, among the sacral-sparing examinations, AIS could be determined only with the S4-5LT, and it might be possible to obtain similar results as those obtained with the ISNCSCI algorithm. The generalizability of our study can aid in determining AIS in individuals with TSCI within three months after injury.

The originality of this study is that all the items in the sacral-sparing examination during the change from complete to incomplete paralysis were studied in detail from the injury up to three months. In addition, among these examinations, the S4-5LT was shown to be a reliable method for determining complete or incomplete paralysis due to its high discriminatory power.

In clinical practice, simplification of the evaluation items will reduce the burden on the examiner as well as the individual. The VAC and DAP examinations may cause reflexive defecation, which is inconvenient for both the examiner and the individual and may cause the person to reject these examinations. People tend to prefer a simpler and less invasive evaluation. Against this background, Marino et al. reported that S3 pressure sensation is a reliable alternative to DAP and other tests.⁹ Our results may also be a valuable source of information because ASIA has recently devised ways to simplify measurement using the ISNCSCI worksheets, which consist of 134 tests depending on specific situations.²¹

This study had some limitations. First, as this study design was a retrospective cohort study, the results of this study were only suggestive and did not provide a high level of evidence. Second, the recovery of patients with TSCI was evaluated for three months after injury, excluding patients with lumbar levels and those that recovered at later time points.

Conclusions

Our findings suggested that the S4-5LT was the most reliable and showed superior performance among the sacral-sparing examinations and could produce results similar to those of the AIS determination method using the standard ISNCSCI algorithm.

Disclaimer statements

Acknowledgements None.

Data availability statement The Japan Single Center study for Spinal Cord Injury Database, which was generated and analyzed during the current study, is not publicly available due to the protection of personal information. However, the data are available from the corresponding author on reasonable request.

Conflicts of interest No potential competing interest was reported by the authors.

Author contributions YA was responsible for designing the studies, screening potentially eligible data, analyzing data, interpreting results, creating tables, and drafted the manuscript. TH designed the review protocol, supervised the overall project, and performed the final manuscript preparation. RI, SM, and RI contributed to data extraction and provided feedback on the report. RK and TN were responsible for designing the review protocol and interpreting results. HK, HS, FT, and TM contributed to arbitrating potentially eligible data and interpreting results. All authors read and approved the final manuscript.

Funding None.