Incidence and risk factors of pneumonia following acute traumatic cervical spinal cord Injury

Abstract

Objectives

To elucidate the incidence and risk factors for pneumonia after acute traumatic cervical spinal cord injury (CSCI)

Design

Retrospective cohort study.

Setting: Spinal injuries center in Japan.

Participants: Of 184 individuals who were admitted within 2 weeks after acute traumatic cervical spinal injuries, 167 individuals who met the criteria were included in this study.

Interventions: The occurrence of pneumonia, degree of dysphagia using the Dysphagia Severity Scale, patient age, history of smoking, presence of tracheostomy, vital capacity, level of injury, and the American Spinal Injury Association Impairment Scale (AIS) 2 weeks after injury were assessed.

Outcomes: Incidence of pneumonia were analyzed. Moreover, the risk factors of pneumonia were evaluated using logistic regression analysis.

Results

From the 167 individuals who met the criteria, 30 individuals (18%) had pneumonia; in 26 (87%) of these individuals, pneumonia was aspiration related, defined as Dysphagia Severity Scale ≤ 4. The median occurrence of aspiration pneumonia was 11.5 days after injury. A logistic regression analysis revealed that severe AIS and severe Dysphagia Severity Scale scores were significant risk factors of pneumonia after CSCI.

Conclusions

It was highly likely that the pneumonias following CSCI were related to aspiration based on the Dysphagia Severity Scale. In addition, most of the patients developed aspiration pneumonia within 1 month after injury. Aspiration and severe paralysis were significant risk factors for pneumonia. The treatment of dysphagia in the acute phase should be considered an important indicator to prevent pneumonia.

Introduction

In the United States, diseases of the respiratory system were the leading cause of death in individuals with spinal cord injury. Of these, 65.2% of cases were pneumonia in patients with spinal cord injuries.¹ Frequent respiratory complications following acute cervical spinal cord injury (CSCI) varied according to the level of injury, as individuals with CSCI at C1–C4 had pneumonia (63%), ventilatory failure (40%), and atelectasis (40%), whereas those with CSCI at C5–C8 had atelectasis (34%), pneumonia (28%), and ventilatory failure (23%).² Pneumonia following acute CSCI is associated with the severity of paralysis³ because CSCI affects respiratory muscles and respiratory tract secretions, which may cause atelectasis. In addition, recent studies reported that age at injury, vital capacity, tracheostomy, and smoking were associated with respiratory complications and mortality after spinal cord injury.^{4, 5}

The relationship between CSCI and aspiration pneumonia has not been well understood, although the relationship between CSCI and dysphagia has been reported.⁶ Only a few previous studies demonstrated a higher occurrence of pneumonia in CSCI patients with dysphagia^{7, 8}; however, to the best of our knowledge, no previous study using a multivariate analysis detected dysphagia as a risk factor of pneumonia. The treatment methods should be different in terms of selecting antibiotics, type of food, oral care, and intervention of rehabilitation for dysphagia, whether the pneumonia was associated with aspiration or not. We hypothesized that swallowing dysfunction might be one of the causes of pneumonia following acute CSCI. The purpose of this study was to elucidate the incidence, the period of occurrence, and risk factors for pneumonia following acute traumatic CSCI.

Methods

Participants

A total of 184 consecutive individuals admitted within 2 weeks after acute traumatic cervical spinal injury to the Spinal Injuries Center at Fukuoka prefecture in Japan between October 2015 and August 2019 were included in this study. We retrospectively reviewed the medical records in the database system (Japan Single Center Study for Spinal Cord Injury Database)⁹ and extracted the data. The exclusion criteria were as follows: individuals with other diseases that caused dysphagia (cerebral infarction, brain injury, and other neuromuscular diseases), those who had difficulty communicating in a physical examination because of dementia and congenital disease, and those who switched to another hospital because of deterioration in their general conditions. The institutional review board at our institute approved this study.

Evaluation

Pneumonia was diagnosed as clinically defined pneumonia according to CDC criteria.^{10, 11} The diagnostic algorithm for clinically defined pneumonia comprises combinations of the following timely available clinical and laboratory parameters—At least 1 of the following: leukopenia or leukocytosis, altered mental status in patients ≥ 70 years of age, fever >38 °C; and at least 2 of the following: new onset or changes in purulent sputum or respiratory secretions, new onset of cough/dyspnea, pathological auscultatory findings, worsening gas exchange such as O₂ desaturations, increased O₂ requirements or increased ventilator demand, tachypnea; and ≥2 serial chest radiographs with at least 1 of the following: new or progressive infiltrations, consolations or cavitation in a chest X-ray. In patients without underlying pulmonary or cardiac disease, 1 definitive chest radiograph is acceptable.

Severity of dysphagia was classified using Dysphagia Severity Scale (DSS).¹² The DSS evaluates the condition of aspiration or dysphagia using the following scores: 1, saliva aspiration; 2, food aspiration; 3, water aspiration; 4, occasional aspiration; 5, oral problems; 6, minimum problems; and 7, within normal limits. Patients with a scale from 1 to 4 were categorized as “the presence of aspiration.” We also used fiberoptic endoscopic evaluation of swallowing (FEES) or videofluoroscopy (VFS) to make a detailed evaluation and decide the classification of DSS for patients categorized with a DSS score ≤ 4. DSS scores were discussed and determined by T.H. (first author) and Y.F. (second author) in reference to the FEES or VFS results.

Aspiration pneumonia was diagnosed as pneumonia in patients with a predisposition to aspiration because of dysphagia or swallowing disorders, according to the consensus by the Japanese Study Group of the committee of Japan Respiratory Society on Aspiration Pulmonary Disease.¹³ We defined aspiration pneumonia as pneumonia with a DSS score ≤ 4 for this study.

The American Spinal Injury Association (ASIA) Impairment Scale (AIS)¹⁴ and vital capacity using portable spirometry⁴ at 2 weeks after injury were also evaluated by both orthopedic surgeons and trained physical therapists. The level of injury, which means skeletal level, was detected by two surgeons using MRI and CT. Age, sex, history of smoking, and the presence of tracheostomy were also assessed.

Data analyses

The individuals with pneumonia were compared with those without pneumonia. A chi-square test was used after the variables were categorized. A logistic regression model was used to compute odds ratios (ORs) and 95% confidence intervals (95% CI) for an increased risk of pneumonia.

According to the World Health Organization definition, age was categorized into two groups: ≥ 65 years and < 65 years. The level of injury was categorized into upper (C1–2, C2–3), middle (C3–4, C4–5), and lower (C5–6, C6–7, C7–T1). At 2 weeks after injury, dysphagia was divided into two categories defined by the presence of aspiration: scale 1–4 and scale 5–7.¹² Vital capacity was classified into two groups based on the median because no definitive cutoff points were found in previous articles: >1100 ml and ≤1100 ml.

The logistic regression model was adjusted for age, DSS, AIS, level of injury, vital capacity, history of smoking, and the presence of tracheostomy. All statistical analyses were performed using JMP® 10 (SAS Institute Inc., Cary, NC, USA) computer software. A P-value of less than 0.05 was considered statistically significant.

Results

Of the 184 individuals admitted to our spinal injuries center within 2 weeks after trauma, individuals were excluded due to: another disease that causes swallowing dysfunction, six individuals; cerebral infarction, five; and muscular dystrophy, one; and difficulty in communicating in a physical examination (dementia, two; and down syndrome, one). Moreover, eight others were excluded because they shifted to another hospital due to the deterioration of their general condition. The remaining 167 individuals were included in this study (Figure 1). Table 1 summarizes their demographic characteristics.

Figure 1.

Study flowchart.

Table 1.

Demographic data of this study.

	Inclusion group	Exclusion group	P-value†
	(n = 167)	(n = 17)
Age (years), n (%)	63.5 ± 16.3	70.1 ± 17.3	n.s.
Sex (male), n (%)	128 (77)	15 (88)	n.s.
Dysphagia Severity Scale	5 (4–6)	N/A
ASIA Impairment Scale, n (%)
A	40 (24)	N/A
B	18 (11)	N/A
C	42 (25)	N/A
D	48 (29)	N/A
no paresis	19 (11)	N/A
Vital Capacity (ml)	1178 ± 642	N/A
History of smoking (positive), n (%)	111 (66)	10 (59)	n.s.
Tracheostomy (positive), n (%)	17 (10)	3 (18)	n.s.

Abbreviations: ASIA, American Spinal Injury Association; N/A, not applicable; n.s., not significant

Values are given as number of patients (%), mean ± standard deviation, or median (interquartile range)

†P-value was caluculated by the Student's t-test or the χ²-test.

Of the 167 individuals who met the criteria, 30 individuals had pneumonia. In these 30, 26 individuals, who had DSS scores from 1 to 4, were diagnosed with aspiration pneumonia. Therefore, pneumonia and aspiration pneumonia incidences after traumatic CSCI were 18% and 16%, respectively. The ratio of aspiration pneumonia to pneumonia was 87% (26/30 individuals).

The median number of days when both pneumonia and aspiration pneumonia occurred after injury were 10 days (range: 3–250 days) and 11.5 days (range: 3–88 days), indicating that these cases of pneumonia occurred at the acute phase of CSCI. Of the 30 CSCI individuals who experienced pneumonia, 24 individuals (80%) developed pneumonia within 1 month after injury (Figure 2). Similarly, of the 26 individuals who were diagnosed with aspiration pneumonia, 22 individuals (85%) developed aspiration pneumonia within 1 month after injury (Figure 3).

Figure 2.

Period of the occurrence of pneumonia.

Figure 3.

Period of the occurrence of aspiration pneumonia.

CSCI patients with and without pneumonia were compared in Table 2. The χ² test revealed that those with pneumonia had significantly severe dysphagia, severe AIS, lower vital capacity, and presence of tracheostomy. After adjustment for potential confounding factors, significantly elevated ORs were observed in individuals with DSS scores ≤ 4 (OR: 5.89, 95% CI: 1.64–21.2), and AIS A or B (OR: 5.56, 95% CI: 1.61–19.2), indicating that severe dysphagia and severe paralysis were significant risk factors for pneumonia following acute traumatic CSCI.

Table 2.

Risk factors for pneumonia in acute cervical spinal cord injury.

	Pneumonia		P-value†	Crude OR (95% CI)	Adjusted OR (95% CI)
	(+) n = 30	(–) n = 137
Age (years), n (%)
<65	11 (37)	57 (42)		Reference	Reference
≥65	19 (63)	80 (58)	0.618	1.23 (0.54–2.78)	1.17 (0.36–3.79)
Dysphagia Severity Scale, n (%)
≥5	5 (17)	107 (78)		Reference	Reference
≤4	25 (83)	30 (22)	<0.001	17.83 (6.29–50.56)*	5.89 (1.64–21.17)*
ASIA Impairment Scale, n (%)
C, D	6 (20)	103 (75)		Reference	Reference
A, B	24 (80)	34 (25)	<0.001	12.11 (4.57–32.13)*	5.56 (1.61–19.22)*
Level of injury, n (%)
C1–2, C2–3	2 (7)	16 (12)		Reference	Reference
C3–4, C4–5	21 (70)	68 (50)		2.47 (0.52–11.63)	1.24 (0.13–11.91)
C5–6, C6–7, C7–T1	7 (23)	53 (39)	0.129	1.06 (0.19–5.60)	0.53 (0.05–5.79)
Vital Capacity, n (%)
>1100 ml	6 (20)	70 (51)		Reference	Reference
≤1100 ml	24 (80)	67 (49)	0.002	4.18 (1.61–10.86)*	1.47 (0.42–5.05)
History of smoking, n (%)
Negative	6 (20)	50 (36)		Reference	Reference
Positive	24 (80)	87 (64)	0.08	2.29 (0.88–6.00)	2.58 (0.73–9.07)
Tracheostomy, n (%)
Negative	18 (60)	132 (96)		Reference	Reference
Positive	12 (40)	5 (4)	<0.001	17.6 (5.55–55.79)*	2.75 (0.67–11.22)

Abbreviations: ASIA, American Spinal Injury Association; CI, confidence interval; OR, odds ratio

†P-value was calculated by the χ²-test.

*P < 0.05 by univariate or multivariate logistic analysis

In addition, of the 167 individuals who met the criteria, 8 individuals (5%) underwent ventilatory support, 165 individuals (99%) ate orally, and 20 individuals (12%) used feeding tube; whereas, of the 30 individuals with pneumonia, 6 individuals (20%) required ventilator support, 28 individuals (93%) ate orally, and 2 individuals (7%) used feeding tube before the occurrence of pneumonia. As the DSS score was evaluated again when pneumonia was diagnosed, in addition to the regular evaluations at two weeks, one month, two months, and three months after injury, there was no interval between the determination of the DSS score and pneumonia diagnosis. No patient in this study received intravenous sedation when we diagnosed pneumonia.

Discussion

In this study, the incidence of pneumonia and aspiration pneumonia was 18% and 16%, respectively, indicating that the ratio of aspiration pneumonia to pneumonia was 87%. Reported incidence rates of pneumonia following spinal cord injury were dependent on diagnostic criteria and had a large range (11–84%).^{2, 3, 15} However, no previous study showed the ratio of aspiration pneumonia to pneumonia. Stroke-associated pneumonia also had a large incidence range (4.1–56.6%) depending on different settings because the pathophysiology of stroke-associated pneumonia is likely explained by aspiration combined with stroke-induced immunodepression.¹⁶ In this study, aspiration pneumonia in individuals with CSCI, in which the relationship between pneumonia and dysphagia in previous studies has not been well examined, was revealed to occur frequently.

The median period for the occurrence of pneumonia and aspiration pneumonia was 10 and 11.5 days after injury, respectively, in this study. Our results, suggesting that pneumonia tended to occur in the acute phase after injury, were consistent with previous studies.^{2, 17} Moreover, aspiration pneumonia often occurs in the acute phase because the dysphagia following CSCI is most severe in the acute phase, which gradually recovers at least 3 months after injury.¹⁸

Severe dysphagia (DSS score ≤ 4), categorized as aspiration, was associated with 5 times the risk of pneumonia compared with mild dysphagia (DSS score ≥ 5) in this study. Although a few studies have reported the relationship between dysphagia and pneumonia in individuals with CSCI, Chaw E and Shem K^{7, 8} reported individuals with dysphagia had statistically higher occurrences of pneumonia when compared with those without dysphagia. Furthermore, multivariate analysis in our study suggested that aspiration was one of the significant risk factors for pneumonia. We previously demonstrated that one of the significant causes for dysphagia following acute CSCI was the morphological change accompanied with swelling of retropharyngeal space,⁶ which might cause pharyngeal residue. Accordingly, pharyngeal residue may cause aspiration pneumonia. As in stroke-associated pneumonia, aspiration played a vital role in the pathogenesis of CSCI-associated pneumonia.¹⁶

Severe motor paralysis (AIS A or B) was also a significant risk factor of pneumonia, whereas the level of injury detected by MRI and/or CT was not a significant predictive factor. Our results were consistent with previous papers,^{3, 19} indicating people with severe paralysis, which tended to incur a respiratory complication. Since complete motor paralysis involves flaccid paralysis of the abdomen and intercostal muscles that leads to weak cough, hypersecretion, and lower tidal volumes,³ these conditions would be associated with the occurrence of pneumonia following acute CSCI.

Some limitations of our study should be acknowledged. Although a previous study² showed that cases of higher level cervical injuries have higher rates of respiratory issues than case of lower level cervical injuries, the level of injury was not a significant risk factor for pneumonia after CSCI in our study. Although this might result from the uneven distribution of severity of paralysis, the severity of paralysis had a stronger effect than the level of injury in this study. In addition, because this was a retrospective study, some factors might be missing in our analyses. Thirty individuals with low DSS score (DSS≤4) did not have pneumonia. Because infection would be associated with the immune status of the patients, aspiration does not always cause pneumonia; however, this was not assessed in this retrospective study. Future studies, including the immune status of the patients, will be needed to clarify this issue.

In conclusion, it was highly likely that the pneumonias following CSCI were related to aspiration based on the DSS score. In addition, most of the patients developed pneumonia within 1 month after injury. Aspiration and severe paralysis were significant risk factors for pneumonia. Regarding clinical practice, our study suggests that the assessment of dysphagia played a critical role when individuals with CSCI experienced pneumonia in the acute phase because the majority of pneumonia was associated with dysphagia associated with aspiration. Changing the type of food eaten, early oral care intervention, and rehabilitation for swallowing dysfunction, which are the basic treatment for dysphagia, may help prevent pneumonia associated with CSCIs.

Data archiving

The datasets generated during and/or analyzed during the current study are not publicly available from the viewpoint of personal information protection but are available from the corresponding author upon reasonable request.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Funding Statement

The works of Dr. Hayashi were supported by JSPS KAKENHI Grant Numbers 19K19861, and a medical research grant on traffic accidents from The General Insurance Association of Japan.

Conflict of interest

No potential conflict of interest was reported by the author(s).

Statement of ethics

The study protocol was approved by the Institutional Review Board of the Spinal Injuries Center, and all participants provided written informed consent before participating in the study. We certify that all applicable institutional and governmental regulations concerning the ethical use of human volunteers were followed during this research.

Patient consent for images or information used in the manuscript

Not applicable.

Authors’ contributions

TH was responsible for designing the study, writing the protocol and report, extracting and analyzing data, screening potential eligible studies, interpreting results, updating reference lists, and creating the “Summary of Findings” tables. YF was responsible for conducting the literature search, extracting and analyzing data, and screening potential eligible studies. OK contributed to interpreting the results. YY contributed to designing the study and interpreting the results. KK contributed to designing the study and interpreting the results. HS contributed to designing the study. MM contributed to interpreting the results. YM contributed to interpreting the results. KK contributed to interpreting the results. KY contributed to interpreting the results. HK contributed to designing the study and interpreting the results. TM contributed to interpreting the results and providing feedback on the report.