Background
First described by Smith and Robinson in 1955 and later modified by Southwick and Robinson the anterolateral approach has become a widely used access to the cervical spine [1, 2]. Anterolateral skin incision, either transverse along the skin crease or longitudinal parallel anterior of the left sternocleidomastoid muscle allows access to the cervical and upper thoracic spine. Once the platysma is divided, blunt dissection between medial infrahyoid muscles and large neurovascular structures lateral leads directly to the spinal column [3].
Indications for surgery vary broadly, including traumatic, degenerative, inflammatory, and neoplastic pathologies. Especially acute cervical spinal cord injury (cSCI) often requires early surgical stabilization. These patients often require intubation and prolonged mechanical ventilation due to neurological impairment. Up to 67% of the affected patients develop respiratory complications ranging from simple atelectasis, pneumonia, and aspiration to respiratory failure, being the most likely cause of death [4, 5]. One-fifth of these patients will require tracheostomy. Age > 50 years, injury severity score (ISS) > 16, neurological level of injury above C5, intubation on scene, high injury burden, and facial and thoracic trauma are independent risk factors for the need of a tracheostomy [6, 7]. Advantages of a tracheostomy in comparison to tracheal intubation include patients’ comfort, a reduction in dead space, decreased work of breathing, reduction of sedatives, and most important the facilitation of the weaning process [8, 9].
Anzeige
Percutaneous dilatational tracheostomy (PDT) has become a standard procedure in intensive care units (ICU) that offers some advantages compared to surgical tracheostomy procedures, including reduction of surgical trauma and operating time. Development of elaborate instruments including single-step [10], balloon [11], forceps [12], or screw-driven [13] dilators, enhanced the implementation of PDT. Preoperative ultrasound and periprocedural bronchoscopy further increased the procedure’s safety.
Historically injuries of the cervical spine have been seen as absolute or relative contraindications for PDT, although evidence was lacking [14‐16]. Possibly this was due to difficulties to identify anatomical landmarks, limited mobility of the neck, and the risk of deteriorating neurologic disability.
Overall morbidity rates after anterior cervical spine fixation (ACSF) surgery are low. However, the unique vascular, neural and aerodigestive anatomy of the anterior neck bears a wide variety of potential complications, including postoperative dysphagia, hardware failure, nerve palsy, cerebral spinal fluid leak, esophageal perforation, and surgical site infections (SSI) [17].
Theoretically, the close proximity of incisions from ACSF and PDT might increase the risk of SSI. However, only few studies investigated the incidence of SSI. Overall incidence rates of SSI ranged from 0 to 5.9% but included infections of the anterior and/or the posterior surgical approach [18‐25]. To prevent SSI some authors recommended a delayed tracheostomy beyond day 14 after ACSF when the surgical site of ACSF is consolidated [8]. This assumption was disproved by several retrospective studies comparing SSI in early and late tracheostomies. All studies favored early tracheostomy, with lower rates of SSI independent of how early and late were defined [19, 21, 25‐27]. Even a preexisting tracheostomy did not increase the risk for SSI after ACSF [22]. Studies including both PDT and surgical tracheostomies are inconclusive with regard to SSI rates using either technique [21, 23‐25, 27].
Anzeige
Complications from tracheostomy other than SSI at ACSF site are largely incompletely reported, and focus on stomal infections, stomal inflammation (i.e. absence of bacterial pathogens), and postprocedural bleeding [18, 20, 23, 24] or intraprocedural complications [25]. Putensen et al. offer straight definitions of all intraoperative, early- and late-postoperative complications of PDT [28]. Initially implemented in a cohort of patients undergoing elective general surgery Clavien-Dindo’s (CD) classification helps to grade the impact of a complication by assessing the therapy to treat the same complication [29]. The purpose of our study was to evaluate the risk of periprocedural complications and SSI of Ciaglia-single step PDT in patients after ACSF and compare these findings to patients receiving PDT for other reasons.
Methods
The aim of this study is to prove the safety of PDT after ACSF. This is a retrospective observational cohort study approved by the local Institutional Ethics Board (BO-EK 87032016) and complied with the declaration of Helsinki. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline was used to ensure proper reporting of methods, results, and discussion. Adult patients after ACSF who received PDT between 2009 and 2020 at the surgical intensive care unit of a tertiary referral hospital were compared to such patients without ACSF.
Demographic data, medical records, surgical procedures, and clinical outcomes were retrieved from patients’ electronic medical files.
Statistical analyses were performed using IBM SPSS Statistics version 28.0.1.0 (IBM Corporation, Armonk, NY). Propensity score matching was performed by one-to-one matching without replacement. Propensity score was obtained by logistic regression with a caliper width of 0.20. Included covariates were age, BMI, length of ICU stay, and length of follow-up. Each exposed patient (PDT with ACSF) was matched to one unexposed patient (PDT without ACSF). The dataset analysed during the current study are available from the corresponding author on reasonable request.
Continuous data were described as means and analyzed using Mann-Whitney-U-test. Categorical variables were presented as percentages and compared using the chi-square test or Fisher’s exact test. P-values ≤ 0.05 were considered as significant.
PDT was performed using a single-step dilatator as described elsewhere [10] with the slight modification that blunt forceps dissection of the pretracheal tissue as originally described was avoided. Prior to the PDT the patient’s neck underwent a thorough examination, including a bedside ultrasound. Anatomical landmarks were identified, and the surgical site of previous cervical spine surgery was considered. Neck hyperextension was strictly avoided, and any orthosis was removed. PDT was performed under permanent bronchoscopic visualization using a sterile single-use Ciaglia Blue Rhino set (Cook Medical, Bloomington, Indiana) and a Shiley tracheostomy tube (Covidien, Mansfield, Massachusetts). Following the tracheostomy tube placement, the posterior tracheal wall was bronchoscopically examined to rule out injuries.
Results
Baseline characteristics of included individuals
During the study period, 1175 patients received a Ciaglia single-step PDT in our surgical ICU. Fifty-seven patients were identified who underwent PDT after cervical spine surgery using an anterior approach. After propensity matching, these patients were compared to fifty-seven control subjects. Baseline characteristics are shown in Table 1. Between groups, age, gender, length of ICU stay, length of follow-up and intra-hospital mortality did not differ significantly. For specific characteristics of the ACSF group, see Table 2. In the ACSF group, median interval from cervical spine surgery to PDT was 11.3 days. In patients receiving exclusive anterior corpectomy, grafting with bone or vertebral body substitute and plating (n = 40) the interval was 10.5 days, in patients with additional posterior fixation (n = 13) 12.5 days and in patients with anterior odontoid screw fixation (n = 3) 17.0 days. During the follow-up period seven patients in the ACSF group and eleven patients in the non-ACSF group died for reasons unrelated to PDT.
Table 1
Patient with PDT with or without ACSF (baseline characteristics). Values in parentheses are percentages unless indicated otherwise. * values are mean (SD). ACSF anterior cervical spine fixation, BMI body mass index, ICU intensive care unit, n.a. not applicable, PDT percutaneous dilational tracheostomy
Baseline characteristics | |||
|---|---|---|---|
ACSF patients (n = 57) | non-ACSF patients (n = 57) | p value | |
Age (years)* | 65.2 (18.0) | 66.2 (15.1) | 0.740 |
BMI (kg/m2)* | 25.9 (3.2) | 26.0 (4.4) | 0.963 |
Gender (male, %) | 49 (86.0) | 37 (64.9) | 0.009 |
Length of ICU stay (days)* | 26 (19) | 24 (11) | 0.464 |
Length of Follow-up (days)* | 388 (791) | 424 (819) | 0.405 |
Interval ACSF and PDT (days)* | 11.3 (7.8) | n.a. | n.a. |
PDT under concomitant antibiotic treatment for other reasons (n, %) | 40 (70.2) | 44 (77.2) | 0.395 |
Intra-hospital mortality (n, %) | 3 (5.2) | 8 (14.0) | 0.113 |
Anzeige
Table 2
Clinical data of ACSF patients. Values in parentheses are percentages unless indicated otherwise. * values are mean (SD). ACSF anterior cervical spine fixation, BMI body mass index, MVA motor vehicle accident, TBI traumatic brain injury
Characteristics of ACSF patients (n = 57) | |
Etiology | |
traumatic | 51 (89.5%) |
infectious | 4 (7.0%) |
neoplastic | 2 (3.5%) |
Ankylosing spondylitis | 5 (8.8%) |
Surgical procedure | |
Exclusive anterior cervical spine fixation | 41 (71.9%) |
Exclusive anterior odontoid screw fixation | 3 (5.3%) |
Additional posterior fixation | 13 (22.8%) |
Trauma mechanism (n = 51) | |
Jump/Fall > 2 m | 9 (17.6%) |
MVA | 10 (19.6%) |
Fall on head | 13 (25.5%) |
Fall over stairs | 10 (19.6%) |
Bicycle accident | 4 (7.8%) |
Brawl | 1 (2.0%) |
Jump/Fall in flat water | 3 (5.9%) |
Hit by falling item | 1 (2.0%) |
concomitant injuries | |
TBI | 20 (39.2%) |
Face | 9 (17.6%) |
Chest | 21 (41.2%) |
Abdomen | 9 (17.6%) |
Extremities | 14 (27.5%) |
Intra- and postprocedural complications of PDT
When Putensen’s definitions of complications of PDT were applied, eleven (19.3%) patients in the ACSF group had at least one complication, compared to twelve patients (21.1%) in the non-ACSF group, with no significant difference between groups (p = 0.815). In each group, one patient had more than one complication (n.s.).
No intraprocedural complications from PDT were reported in the ACSF group, while three patients in the non-ACSF group (5.3%) had intraprocedural complications. Specifically, two injuries of the tracheal wall requiring surgical intervention (one defined as pneumomediastinum with pneumothorax and the other as “false passage”) occurred. Furthermore, one difficult insertion of the tracheal cannula occurred in this group due to material failure, which was solved when a different tracheostomy kit was used. These differences in both groups did not reach statistical significance (p = 0.079).
The majority of postoperative complications comprise mild local inflammation (8.8% vs. 3.5%) and minor bleeding (7.0% vs. 8.8%) at the stomal site. Differences in both groups again were not statistically significant (p = 0.809).
Anzeige
Loss of airway is a heterogenous term that can describe a range of clinical severity. In the ACSF group, one patient required resuscitation on the 9th postoperative day when the cannula was completely obstructed with mucus. After 4 min, return of spontaneous circulation was achieved. Another patient needed uneventful cannula change due to obstruction. In the non-ACSF group, self-inflicted removal of the tracheostomy cannula occurred in two patients. One patient required a reinsertion of the cannula under general anesthesia, while in the other patient the cannula was replaced bedside. Table 3 summarizes all detected complications in the study cohort.
Table 3
Incidence of complications using Putensen’s definitions. ACSF anterior cervical spine fixation
ACSF patients (n = 57) | non-ACSF patients (n = 57) | p value | |
|---|---|---|---|
Complication (events) | |||
Cuff leakage | 1 (1.8%) | 1 (1.8%) | 1.000 |
Difficult dilatation | 0 | 0 | 1.000 |
Difficult insertion | 0 | 1 (1.8%) | 0.315 |
Esophageal perforation | 0 | 0 | 1.000 |
False passage | 0 | 1 (1.8%) | 0.315 |
Gastric aspiration | 0 | 0 | 1.000 |
Hypotension | 0 | 0 | 1.000 |
Hypoxemia | 0 | 0 | 1.000 |
Inflammation | 5 (8.8%) | 2 (3.5%) | 0.242 |
Infection | 0 | 0 | 1.000 |
Minor bleeding | 4 (7.0%) | 5 (8.8%) | 0.728 |
Major bleeding | 0 | 1 (1.8%) | 0.315 |
Loss of airway | 2 (3.5%) | 2 (3.5%) | 1.000 |
Pneumothorax | 0 | 1 (1.8%) | 0.315 |
Pneumomediastinum | 0 | 1 (1.8%) | 0.315 |
Subcutaneous emphysema | 0 | 0 | 1.000 |
Tracheo-innominate fistula | 0 | 0 | 1.000 |
Complication of PDT requiring reoperation (including surgical hemostasis) | 2 | 3 | 0.633 |
PDT-related mortality | 0 | 0 | 1.000 |
Number of complications per patient | |||
no complication | 46 (80.7%) | 45 (78.9%) | 0.815 |
single complication | 10 (17.5%) | 11 (19.3%) | 0.809 |
two complications | 1 (1.8%) | 1 (1.8%) | 1.000 |
Patients with one or more than one complication | 11 (19.3%) | 12 (21.1%) | 0.815 |
Intra- or postoperative complications (patients) | |||
intraoperative complications | 0 | 3 (5.3%) | 0,079 |
postoperative complications | 11 (19.3%) | 10 (17.5%) | 0,809 |
Due to the inherent limitations of Putensen’s complication definitions, we additionally compared Clavien-Dindo’s (CD) classification of surgical complications score to include a broader spectrum of clinically relevant complications between groups and results were stratified into none or mild, moderate and severe complications (Table 4). Overall, no significant differences were detected between the two groups. Furthermore, the general incidence of life-threatening complications was very low (1.8% vs. 3.5%, p = 0.558).
Table 4
Incidence of complications using Clavien-Dindo’s classification. ACSF anterior cervical spine fixation, CD Clavien-Dindo’s classification
Clavien-Dindo classification (patients) | |||
|---|---|---|---|
ACSF patients (n = 57) | non-ACSF patients (n = 57) | p value | |
0 | 46 (80.7%) | 45 (78.9%) | 0.815 |
I | 7 (12.3%) | 7 (12.3%) | 1.000 |
IIIa | 3 (5.3%) | 1 (1.8%) | 0.309 |
IIIb | 0 | 2 (3.5%) | 0.154 |
IVa | 1 (1.8%) | 2 (3.5%) | 0.558 |
no or mild complications (CD 0 – II) | 53 (93%) | 52 (91.2%) | 0.728 |
moderate complications (CD III) | 3 (5.3%) | 3 (5.3%) | 1.000 |
severe complications (CD IV – V) | 1 (1.8%) | 2 (3.5%) | 0.558 |
Anzeige
Surgical site infections
None of the patients in the ACSF group showed inflammation or infection of the surgical site of ACSF. Two patients required reoperation due to implant failure not related with an infection. One patient presented with cervical swelling caused by cerebrospinal fluid leak because of a dural tear. It was successfully treated with a temporary external lumbar drainage. Two patients in the ACSF group had infections of their posterior surgical access to the cervical spine requiring surgical revision. These infections were unrelated to the tracheostomy since one occurred prior to PDT, and the other was clearly related to a collar-induced pressure ulcer.
Discussion
This study compared intra- and postprocedural complications of Ciaglia-single step PDT of fifty-seven patients after ACSF to fifty-seven non-ACSF after propensity score matching.
Current knowledge of complications of PDT after ACSF is limited to one prospective trial, and several retrospective case reports and case series. If we add our study to the currently available literature [18‐21, 24, 25, 30‐32], we were able to identify 346 patients receiving a PDT after ACSF (Table 5). Most frequently Ciaglia single-step technique was used for PDT (n = 253, 73.1%), followed by modified Griggs technique (n = 48, 13.9%). Remarkably not all studies report the PDT technique used in their series [18, 24, 33]. In our cohort, we identified 12 complications in 11 patients. The majority comprised local stomal inflammation with an incidence of 8.8% and minor parastomal bleeding (7.0%). This is consistent with previous studies. Reported rates range from 0 to 8.3% in patients after ACSF for stomal inflammation [18, 24, 32] and 0–12.0% for minor bleeding [18, 31, 32]. Loss of airway occurred in two patients (3.5%) due to cannula obstruction. Whether these events are directly related to PDT is debatable. Nevertheless, we decided to assess them as PDT-related complication, because removal of an obstructed cannula might lead to a direct closure of a PDT wound, making it impossible to insert a new cannula, thereby aggravating this potentially life-threatening complication.
Table 5
Studies evaluating complications and the risk of SSI following PDT and ACSF. * Surgical tracheostomies were excluded from this overview, ** this study included patients with CSI with and without ASCF. ACSF anterior cervical spine fixation, CSI cervical spine injury, mo months, d days, n.r. not reported, PDT percutaneous dilational tracheostomy, SSI surgical site infection
Author | Year | Study design | n | SSI | PDT complications | Follow-up | PDT procedure |
|---|---|---|---|---|---|---|---|
Mazzon [30] | 1996 | case report | 1 | 0 | none | 10 d | Ciaglia multi step |
Šustić [36] | 1999 | case report | 2 | n.r. | none | 1 mo | dilational forceps |
Šustić [31] | 2002 | prospective trial | 8* | n.r. | 1 prolonged bleeding | dilational forceps | |
O’Keefe [24] | 2004 | retrospective case series | 12* | 0 | 1 cellulitis | 14 mo | n.r. |
Ben Nun [32] | 2006 | retrospective case series | 38 | n.r.** | 1 minor bleeding 1 cellulitis | 18 mo | dilational forceps |
Romero-Ganuza [18] | 2011 | retrospective case series | 28 | 0 | 1 minor bleeding 2 cellulitis no pneumothorax, pneumomediastinum, tracheal perforation, tracheoesophageal fistula were observed | n.r. | n.r. |
Babu [25] | 2012 | retrospective case series | 4* | 0 | none | 12.5 mo | n.r. |
Kaczmarek [20] | 2017 | retrospective case series | 51 | 0 | 1 infection no accidental decannulation, cardiopulmonary complication, no life-threatening complications | 6 mo | Ciaglia single step |
Lozano [21] | 2018 | retrospective case series | 77* | 1 | 1 esophagocutaneous fistula | n.r. | Ciaglia single step |
Shan [19] | 2020 | retrospective case series | 68 | 0 | 1 infection | n.r. | Ciaglia single step |
current study | 2024 | retrospective case series | 57 | 0 | see main text | 12 mo | Ciaglia single step |
The shortcomings of the majority of the compared studies are incomplete information about the complications that were under investigation. Besides our study, only Kaczmarek et al. [20] and Romero-Ganuza et al. [18] present a broader overview of complications, even if they were not detected in their study cohorts.
After propensity score matching, we were able to compare our results to patients without prior cervical surgery. Importantly, no significant differences in complication rates and severity were detected.
We compared our findings to studies evaluating complications of Ciaglia single-step PDT in a general patient cohort. Yaghoobi et al. reported an overall complication rate of 40%, with minor bleedings being the most common. The authors classified the majority of complications as minor and quickly improvable. Only one potentially life-threatening complication (loss of airway) was seen [34]. A current meta-analysis reports an overall rate of technical difficulties of 7.3% in 716 PDTs independent from the PDT technique used [35]. Those rates are comparable to the ones we found in our ACSF patient cohort, further corroborating the feasibility and safety of PDT in patients after ACSF.
Loss of airway, false passage, tracheal or esophageal injury, major bleeding, pneumothorax, and pneumomediastinum are potentially life-threatening events that require special consideration. Klotz et al. found a rate of 8.7% life-threatening events (81 events) in 926 PDTs in the meta-analysis already mentioned above [35]. To eliminate a subjective interpretation and to avoid under- or overgrading of complications, we used the CD classification. It is a well-established and widely used score, well suited for the interpretation of retrospective data. Hereby, we were able to identify one single life-threatening event (loss of airway followed by successful resuscitation) in the ACSF group (1.8%) and two life-threatening events (tracheal wall injuries requiring surgical interventions) in the non-ACSF group (3.6%).
In our study, we did not observe any SSI during the follow-up period of more than one year. This is consistent with previous studies. Only Lozano et al. identified one case of SSI following PDT due to an esophagocutaneous fistula related to an esophageal perforation by implant material and highly likely not related to PDT [21]. These data suggest that PDT-driven SSI in patients after ACSF are negligible.
A limitation of this study is the retrospective single-center character. Moreover, it does not compare open and percutaneous tracheostomies or different PDT techniques in patients after ACSF. Furthermore, due to the high number of patients with traumatic brain injury in our cohort, valid statements about neurologic improvement or deterioration cannot be given.
Conclusion
Based on our data, we conclude that Ciaglia single-step PDT is a safe and feasible method in patients after ACSF. We strongly recommend that cervical spine injury (CSI) and ACSF be no longer considered as a contraindication for PDT and encourage intensivists to perform urgently needed PDT earlier as benefits for patient’s respiratory recovery are obvious. SSI related to PDT are negligible and the entire risk profile does not differ from non-ACSF-patients. CD classification might help to improve the assessment of PDT-related complications. However, this procedure should be performed after a thorough examination of the operating field, including the use of ultrasound, and permanent bronchoscopic visualization by or under the supervision of experienced intensive care specialists.
Declarations
Conflict of interest
The authors declare that they have no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
