Risk Analysis Based on the Timing of Tracheostomy Procedures in Patients with Spinal Cord Injury Requiring Cervical Spine Surgery

Accepted Manuscript Risk analysis based on the timing of tracheostomy procedures in patients with spinal cord injury requiring cervical spine surgery Rita Galeiras, MD, PhD, Mónica Mourelo, MD, María Teresa Bouza, MD, María Teresa Seoane, PhD, María Elena Ferreiro, MD, Antonio Montoto, MD, Sebastián Salvador, MD, Leticia Seoane, MD, David Freire, MD PhD PII:

S1878-8750(18)31018-0

DOI:

10.1016/j.wneu.2018.05.065

Reference:

WNEU 8126

To appear in:

World Neurosurgery

Received Date: 24 February 2018 Revised Date:

9 May 2018

Accepted Date: 10 May 2018

Please cite this article as: Galeiras R, Mourelo M, Bouza MT, Seoane MT, Ferreiro ME, Montoto A, Salvador S, Seoane L, Freire D, Risk analysis based on the timing of tracheostomy procedures in patients with spinal cord injury requiring cervical spine surgery, World Neurosurgery (2018), doi: 10.1016/j.wneu.2018.05.065. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT TITLE Risk analysis based on the timing of tracheostomy procedures in patients with spinal cord injury requiring cervical spine surgery AUTHORS Rita Galeiras MD, PhD1 ([email protected]), María Teresa Bouza MD1 ([email protected]), María Teresa Seoane PhD2 ([email protected]), María Elena Ferreiro MD3 ([email protected]), Antonio Montoto MD3 ([email protected])

Leticia Seoane MD1 ([email protected])

M AN U

David Freire MD PhD1 ([email protected])

SC

Sebastián Salvador MD3 ([email protected])

RI PT

Mónica Mourelo MD1 ([email protected]),

1

Critical Care Unit,

2

Clinical Epidemiology and Biostatistics Unit,

3

Spinal Cord Injury Unit.

Complexo Hospitalario Universitario de A Coruña (CHUAC), Sergas, Instituto de Investigación

ABSTRACT

TE D

Biomédica de A Coruña (INIBIC), Universidade da Coruña (UDC), As Xubias, 15006. A Coruña, Spain

Introduction: To determine the optimal moment to carry out a tracheostomy in a patient requiring anterior cervical fixation.

EP

Methods: A retrospective observational study was carried out over an 18-year period on 56 patients who had been admitted to the ICU with acute spinal cord injury (SCI), and who underwent a tracheostomy and

AC C

surgical fixation. The sample was divided into two groups: An at-risk group (31 patients, who had undergone a tracheostomy prior to the cervical surgery or <4 days after the procedure), and a not at-risk group (25 patients, who had undergone a tracheostomy >4 day following the fixation surgery). Both a descriptive and a comparative study were carried out. The overall trend of the collected data was analysed using cubic splines (graphic methods). Results: The only infectious complications diagnosed as related to the surgical procedure were infection of the surgical wound in two patients of the not at-risk group (12%) and deep-tissue infection in one patient of the at-risk group (3.2%). During the study period, we identified a tendency towards the conduct of early tracheostomies.

1

ACCEPTED MANUSCRIPT Conclusions: Our results suggest that the presence of a tracheostomy stoma prior to, or immediately after surgery, is associated with a low risk of infection of the cervical surgical wound in instrumented spinal fusion. KEYWORDS: Tracheostomy; Cervical Spine Surgery; Spinal Cord Injury; Infection.

RI PT

This work did not receive financial support.

MANUSCRIPT TEXT INTRODUCTION

SC

Most quadriplegic patients require intubation and mechanical ventilation (MV). A tracheostomy should be performed as early as possible in patients requiring prolonged MV, as the procedure has demonstrated clear benefits in facilitating respiratory management and reducing complications [1,2].

M AN U

In addition to supportive measures, the current surgical treatment for traumatic spinal cord injury (SCI) involves the use of decompression and stabilization techniques. Early surgical treatment seems to be safe from a clinical standpoint, and could be associated with better neurological outcomes according to recent case series3. However, the optimal timing for surgery has not yet been established prospectively or

facility.

TE D

randomly, and it is subject to the feasibility and availability of expert surgical teams in each hospital

The challenge, therefore, is to determine the optimal moment to carry out a tracheostomy in a patient requiring anterior cervical fixation4. The concern in this regard arises from the fact that the tracheostomy

EP

site is located near the area which is often used to perform cervical stabilization surgery. To this day, only two types of retrospective studies shedding some light on this matter have been carried out. On the one

AC C

hand, 5 studies [1,5-8] evaluating the safety of a tracheostomy carried out after anterior cervical fixation surgery and, on the other, a single study9 assessing the risk of infection in anterior cervical surgery in patients who had previously undergone a tracheostomy. It is hypothesized that the safety of both incisions in the anterior region of the neck depends on the sequencing of both procedures, and that the risk is greater when cervical surgery is carried out in the presence of a previous tracheostomy stoma. The presence of a stoma near a surgical wound from the onset of surgery would entail a greater risk than performing the tracheostomy as of the 4th day [7] or 1-2 weeks after the procedure, as proposed by different authors [1,5,6,8,10], at which time the wound has usually already completed the epithelisation phase and the risk of infection is lower.

2

ACCEPTED MANUSCRIPT In our study, we examined patients who had undergone a tracheostomy for any given reason prior to the cervical surgery, conditioned by the patients’ clinical situation and the availability of an expert surgical team, considering this patient population as an at-risk group as a result of having a tracheal stoma on the day of the intervention, or less than 4 days following the procedure. An interval of 4 or more days after surgery and a subsequent cervical surgical approach were considered to be not at-risk conditions (minimal

RI PT

risk of cross-infection), and patients who met them were included as a control group. MATERIAL AND METHODS

Patient population: Retrospective observational study of adult patients (aged > 18 years) with spinal cord

SC

injury above level D1, who had been admitted to the Intensive Care Unit (ICU) between 1998 and 2016, and who had undergone a tracheostomy and required cervical surgery. Of the total of 382 patients admitted to the ICU with the diagnosis of an acute cervical spinal cord injury, 56 met the study’s inclusion

M AN U

criteria.

Parameters: Relevant data were extracted from the patients’ medical records, including demographic variables (age and sex), comorbidities assessed by Charlson’s index, variables related to the spinal cord injury (structural level, ASIA score, motor index score), severity at admission (Glasgow Coma Scale [GCS], Injury Severity Score [ISS], Acute Physiology and Chronic Health Evaluation II [APACHE II]

TE D

[11], Sequential Organ Failure Assessment [SOFA] [12]), variables linked to the need for respiratory support (days of mechanical ventilation and duration of sedation), variables associated with the tracheostomy (technique used, related complications, days elapsed until the decannulation), variables

EP

related to the cervical surgery (type of surgery, surgical wound infection and deep-tissue infection), length of stay (days), mortality during admission and cause of death.

AC C

Protocol: Our centre is a highly specialized hospital whose Spinal Cord Injury Unit is a benchmark in the autonomous community of Galicia, which had an estimated average population of 2,778,875 during the study period. Patients with acute traumatic SCI are admitted to our ICU if they need, or are expected to need, support for a failing organ. The ICU admits patients who require intensive care and treatment around the clock from staff specializing in critical care and Spinal Cord Injuries. Patients receive fluid resuscitation therapy based on their hourly urine output and blood pressure, early enteral nutrition, mechanical ventilation and vasoactive support, if needed. The surgical intervention is carried out in the Traumatology Unit, comprised by a team of expert spinal surgeons. Surgical treatment includes spinal decompression and instrumented arthrodesis, administered as soon as possible. The approach pathway is

3

ACCEPTED MANUSCRIPT anterior and/or posterior. According to the standard protocol, patients receive perioperative prophylaxis with vancomycin (1gr/12h; 2 doses). The decision to intubate the trachea and initiate mechanical ventilation is made by the attending physician according to standard criteria: decreased level of consciousness, need to protect the airway, hypoxemia, hypoventilation, or dyspnoea. Patients are transferred to the Spinal Cord Injury Unit as soon as their critical situation has been resolved.

RI PT

Definitions: The definition of wound and deep-tissue infections was based on the judgment of the physician responsible for treating the patient at that given time, considering relevant clinical data and complementary studies. Termination of the respiratory support was considered the moment when the

SC

patient no longer required any type of support for 48 consecutive hours. The indication for tracheostomy was established on the basis of the difficulty to intubate the patient or the need for prolonged mechanical ventilation. The surgical tracheostomy was performed by otolaryngology specialists, whereas the

M AN U

percutaneous tracheostomy was carried out by caregivers from the ICU, directly at the patient’s bedside and using the Ciaglia Blue Rhino technique13. Complications taken into account were stomal infection and other non-infectious complications. Patient follow-up ended with hospital discharge. Statistical analysis: A descriptive analysis of all variables collected during the course of the study was carried out. Qualitative variables were expressed as a percentage n(%) and quantitative variables were

TE D

expressed as a mean value with a standard deviation, median and range. Either the Chi-Square test or Fisher's Exact Test was used, as needed, to determine which variables were associated with either risk group (at-risk and not at-risk). A comparison of the mean values of the

EP

continuous variables was carried out using Student’s T-test, or the non-parametric Mann-Whitney U test whenever data did not follow a normal distribution (Kolmogorov-Smirnov test).

AC C

Regression models with cubic splines were used, that is, data was adjusted by joining smoothly-merged polynomial functions, generating a continuous curve, which allowed for identifying time reductions, calculated as the difference in the number of days elapsed between the onset of the lesion and the conduct of the tracheostomy throughout the course of the study.

Ethical aspects: The study was carried out in accordance with the principles set out in the Helsinki Declaration and Decree 29/2009 of the Spanish Law regulating the access to a patient’s medical electronic history. Confidentiality was maintained in accordance with the Spanish Data Protection Act (15/1999). The study was carried out with the approval of our hospital’s Ethics Committee. RESULTS

4

ACCEPTED MANUSCRIPT The mean age of our study population was 48.5 years (SD: 20.2), and 85.7% of patients were men. Charlson's age-adjusted comorbidity index at admission was 0.5 (SD: 1.4). The most frequent cause of the spinal cord injury was traffic accidents (48.2%), followed by falls (39.3%), diving (8.9%) and crush injuries (3.6%). Thirty-eight (38) patients (67.9%) had a level C5-C8 lesion. The severity of the spinal cord lesion upon the patient’s admission to the clinic was assessed with the ASIA classification test, with

RI PT

A being the most frequent category (67.9%). Patients’ mean motor index value was 15.81 (SD: 15.51). The mean ISS score was 26.8 (SD: 12.2). As for the severity scales, the mean GCS score was 13.6 (SD: 3.1) at admission and 14.7 (SD: 1.6) at discharge. Moreover, the mean APACHE II score was 11.6 (SD:

SC

6.7), and the total SOFA score at admission and 4 days after the procedure was 5.4 (SD: 3) and 3.7 (SD: 2.1), respectively (Table 1).

M AN U

Aspects related to the need for mechanical ventilation, cervical surgery and a tracheostomy are shown in Table 1. The prevailing cause of the need for respiratory support was respiratory failure (64.3%). The surgical approach of the cervical injury was anterior in 54 patients (96.4%) and posterior in 2 patients (3.6%). The cervical fixations were made at a single level in 6ha3.6% of the cases. The percutaneous tracheostomy technique was the preferred option in most patients (83.9%), and only 9 patients underwent

TE D

a surgical tracheostomy (16.1%). No attempt was made to extubate patients prior to performing such procedure. The decannulation procedure was successful in 33 patients (58.9%). Table 2 shows that both the at-risk and the not at-risk groups had similar demographic, injury and severity characteristics.

EP

The treatment-related characteristics of both groups are set out in Table 3. Seventeen (17) patients of the at-risk group (68%) and 23 (74.2%) of the not at-risk group received corticosteroid therapy based on the

AC C

clinical recommendations in force during the time period during which the injury took place. Anterior surgical fixation was the most widely used approach in both groups, and the two patients who were operated on posteriorly were included in the not at-risk group. The time elapsed between the onset of the injury and the conduct of the fixation surgery differed significantly between both groups, being shorter in the not at-risk group (median of 7 days vs. 18 days; P = 0.001). The following infectious complications related to the surgical procedure were reported: 3 patients (12%) of the not at-risk group developed an infection of the surgical wound and 1 patient (3.2%) of the at-risk group developed a deeptissue infection. Half (50%) of the patients with a paravertebral abscess or an infection of the surgical wound had a multilevel (> 1 level) fixation.

5

ACCEPTED MANUSCRIPT The technique used to carry out the tracheostomy was similar in both groups; however, the time elapsed between the onset of the lesion and the conduct of the tracheostomy did differ between both groups, being significantly lower in the at-risk group compared to the not at-risk group (median of 8 days vs. 18 days; P <0.001). The median time elapsed between both procedures in the not at-risk group was 11 days. No infectious complications of the tracheal stoma were observed in any of the study patients. With regard to

RI PT

non-infectious complications, 1 patient (4.0%) of the not at-risk group experienced bleeding; as opposed to the lack of such reaction in the at-risk group, 2 patients of each group developed stenosis (8.0% in the not at-risk group vs. 6.5% in the at-risk group) and 1 patient of each group developed a granuloma (4.0%

SC

in the not at-risk group vs. 3.2% in the at-risk group).

We verified whether the absolute time elapsed between the onset of the injury and the conduct of the tracheostomy had decreased since the baseline of this patient cohort, using a smoothed linear model

M AN U

(regression model with cubic splines) in which we divided the variable into smoothly-merged segments in all nodes, applying smooth regression models in each one of them. Furthermore, we implemented a natural cubic spline regression model with an inner node. Figure 1 sets out the results of this analysis, showing that the time elapsed between the two events decreased since the cohort baseline, with the greatest decrease having been observed in the 1998-2010 period.

TE D

No significant differences were observed between both groups with respect to the number days of mechanical ventilation, the number of days of sedation, the time elapsed until the decannulation or the time elapsed until the initiation of oral nutrition. Additionally, we found that the mean ICU stay was

EP

similar in both groups (not at-risk group: 35.2 days (SD: 19.4) vs. at-risk group: 32.9 days (SD: 11). The length of stay was greater in the not at-risk group (245.6 days; SD: 125.5) compared to the at-risk group

AC C

(190.8 days; SD 121.9), but without statistically significant differences. Mortality rates were similar in both groups, and in no case was the death attributable to the tracheostomy and/or surgery (Table 3).

DISCUSSION

The most important findings of this study were:

Firstly: No patient in the at-risk group developed an infection of the surgical wound, and only 1 case of a peri/paravertebral abscess was recorded. Such case corresponded to a polytraumatized patient (ISS = 32) with a total SOFA score of 4 at day 4, who had received a course of corticoid therapy. This patient did not receive intraoperative antibiotic prophylaxis despite the Unit’s standard of care protocol and underwent

6

ACCEPTED MANUSCRIPT bone grafting with instrumented fixation at a single level. The postoperative radiological follow-up assessment yielded normal results; however, in the days following the procedure he developed febrile syndrome secondary to a bacteraemia caused by a methicillin-sensitive staphylococcus. A cervical MRI was performed revealing a perivertebral collection in the arthrodesis area, without evidence of spinal cord compression. Both the surgical scar and the percutaneous tracheostomy stoma which had been performed

RI PT

7 days prior to the cervical surgery were in good condition. No signs of endocarditis were observed in the transesophageal echocardiography performed, and no data suggestive of a catheter-related infection were identified either. The patient’s condition resolved after 8 weeks of antibiotic therapy with cloxacillin and

SC

rifampicin, and he did not require surgical treatment.

The risks associated with the presence of a tracheostomy stoma prior to the cervical surgery, or with the

M AN U

conduct of such procedure in the days immediately following surgery, have been scarcely reported in the available literature. To our knowledge, only Northrup et al.[9] published a retrospective study reviewing the clinical course of 11 patients with a tracheostomy stoma before undergoing anterior cervical spine surgery. None of the patients of this case series developed a surgical wound or deep-tissue infection after surgery. Fifty per cent (50%) of cases underwent instrumented fusion with a bone graft. The authors

TE D

recommend thoroughly preparing the skin and performing the incision laterally to the tracheostomy stoma. Our results suggest that the presence of a tracheostomy stoma before or immediately after surgery is linked to a low risk of infection of the cervical surgical wound, based on our case series in which 100% of patients underwent instrumented fixation. The case of the patient who developed a deep-tissue

EP

collection was interpreted as a primary infection of the surgical bed resulting from an accidental breach of

AC C

the antibiotic prophylaxis protocol, and did not appear to be determined by the cervical incisions.

Other studies have addressed the risks associated with the conduct of a tracheostomy after cervical spine surgery. O'Keeffe et al.[6] asserted in a retrospective study involving 17 patients with a cervical SCI that the conduct of a tracheostomy between 6 and 10 days following a cervical fixation procedure was safe. In their case series, Babu et al. [5] reported that a tracheostomy could be performed without complications after a median of approximately 6 days following surgery. Romero-Ganuza [8] suggested applying an average time interval between the fixation surgery and the tracheostomy procedure of 8.25 ± 5.57 days. Berney et al.7 claimed that performing a tracheostomy as early as 4 days after surgery was associated with minimal complications. In our case series, the median time elapsed between both procedures in the not at-

7

ACCEPTED MANUSCRIPT risk group was 11 days. Three patients in the not at-risk group developed a surgical wound infection. These corresponded to cases of spinal cord injury not associated with another trauma (ISS = 25). In two of these patients, the cervical surgical approach was anterior fixation multilevel (>1 level).. Furthermore, one of the patients had received corticosteroid therapy and another one had undergone a surgical tracheostomy. None of these patients developed an infection of the tracheostomy stoma and all cases

RI PT

resolved after administering conservative therapy. No patient in the at-risk group experienced a deeptissue infection. The available literature shows that in 1% of cases, patients develop an infection of the tracheostomy stoma and the surgical wound whenever the procedure is carried out an average of 6.6 days

SC

after surgery.

Secondly: No patient in our series developed a tracheal stoma infection. A similar number of non-

M AN U

infectious complications were recorded with both procedures, although 83.9% of cases accounted for a percutaneous tracheostomy. Such data differs from those set out by Berney et al. and Babu et al. [5,7] in their case series, as most of their patients had undergone surgical tracheostomies. In broad case series, both percutaneous and surgical tracheostomies proved to be safe. Ben Nun et al. [14] found that percutaneous tracheostomies are a feasible and safe procedure in patients with cervical spinal cord injury

TE D

and anatomical reference points available without neck extension. Percutaneous tracheostomies are faster, they minimize damages inflicted to adjacent neck structures and are probably related to fewer late stomal infections [15], which may constitute a significant advantage in patients who underwent anterior fixation of the cervical spine and require prolonged mechanical ventilation. In our study, the median time elapsed

EP

between the onset of the injury and the conduct of the tracheostomy was 8 days in the at-risk group and 18 days in the not at-risk group. This data proves that patients who underwent a preoperative

AC C

tracheostomy benefited from a greater precocity in this procedure.

We also verified that the time elapsed between the onset of the injury and the conduct of the tracheostomy has decreased over the past few decades, in light of the knowledge on the advantages of this procedure. This downward trend in such times is most likely influenced by the clinician’s familiarity with the percutaneous technique, which has led to a greater autonomy of ICU doctors in the management of critical patients.

Also, the time elapsed between the onset of the injury and the conduct of the surgical procedure differed between both groups. Patients who had undergone an early tracheostomy were subject to a greater

8

ACCEPTED MANUSCRIPT surgical delay. This cannot be explained by differences in the type of lesion, and is probably related to the patients’ clinical condition, which is not reflected by the total SOFA scores at admission and 4 days following surgery.

Thirdly: Despite the fact that the tracheostomy procedure was performed earlier in the at-risk group, this

RI PT

did not translate into shorter lengths of stay, less days of mechanical ventilation, or less time elapsed until the decannulation. It also had no impact on the number of days of sedation or on the time elapsed until the initiation of oral nutrition.

the cervical fixation or tracheostomy placement surgery.

SC

The mortality rates of both groups did not differ significantly and no case of death was associated with

M AN U

Study limitations: Our study was somewhat limited by its reliance on data obtained in a single hospital and by its retrospective nature. We studied a small number of patients, but we included all those with a traumatic cervical spinal cord injury requiring a tracheostomy and cervical surgery over a given period of time. These patients were all subject to the same diagnostic criteria and received similar treatments, thus, increasing the study’s internal validity. While the retrospective nature of our study made it vulnerable to

TE D

information biases resulting from inaccurate clinical records and missing data, the fact that the patients were all in a critical condition upon their admission to the hospital resulted in comprehensive clinical and laboratory analysis data.

EP

CONCLUSIONS

Our results suggest that the presence of a tracheostomy stoma prior to, or immediately after surgery, is

AC C

associated with a low risk of infection of the cervical surgical wound in instrumented spinal fusion. Adherence to good clinical practice guidelines during the surgical and perioperative periods can be a decisive factor in this respect.

9

ACCEPTED MANUSCRIPT REFERENCES 1.

Binder H, Lang N, Tiefenboeck TM, Bukaty A, Hajdu S, Sarahrudi K. Tracheostomy following

anterior cervical spine fusion in trauma patients Int Orthop. 2016;40(6):1157-62. 2.

Leelapattana P, Fleming JC, Gurr KR, Bailey SI, Parry N, Bailey CS. Predicting the need for

3.

RI PT

tracheostomy in patients with cervical spinal cord injury. J Trauma Acute Care Surg 2012;73(4):880-4. Ropper AE, Neal MT, Theodore N. Acute management of traumatic cervical spinal cord injury

Pract Neurol. 2015;15(4):266-72. 4.

Sustić A, Krstulović B, Eskinja N, Zelić M, Ledić D, Turina D. Surgical tracheostomy versus

Spine (Phila Pa 1976). 2002;27(17):1942-5; discussion 5.

Babu R, Owens TR, Thomas S, Karikari IO, Grunch BH, Moreno JR, et al. Timing of

M AN U

5.

SC

percutaneous dilational tracheostomy in patients with anterior cervical spine fixation: preliminary report

tracheostomy after anterior cervical spine fixation J Trauma Acute Care Surg. 2013;74(4):961-6. 6.

O'Keeffe T, Goldman RK, Mayberry JC, Rehm CG, Hart RA. Tracheostomy after anterior

cervical spine fixation J Trauma. 2004;57(4):855-60. 7.

Berney S, Opdam H, Bellomo R, Liew S, Skinner E, Egi M, et al. An assessment of early

2008;64(3):749-53. 8.

TE D

tracheostomy after anterior cervical stabilization in patients with acute cervical spine trauma J Trauma.

Ganuza JR, Garcia Forcada A, Gambarrutta C, Diez De La Lastra Buigues E, Merlo Gonzalez

VE, Paz Fuentes F, et al. Effect of technique and timing of tracheostomy in patients with acute traumatic

9.

EP

spinal cord injury undergoing mechanical ventilation. J Spinal Cord Med 2011;34(1):76-84. Northrup BE, Vaccaro AR, Rosen JE, Balderston RA, Cotler JM. Occurrence of infection in

AC C

anterior cervical fusion for spinal cord injury after tracheostomy. Spine (Phila Pa 1976) 1995;20(22):2449-53. 10. 11.

Ball PA. Critical care of spinal cord injury. Spine (Phila Pa 1976) 2001;26(24 Suppl):S27-30. Leng YX, Nie CY, Yao ZY, Zhu X. [Analysis of the risk factors for early death in acute severe

traumatic cervical spinal cord injury]. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 2013;25(5):294-7. 12.

Stein DM, Menaker J, McQuillan K, Handley C, Aarabi B, Scalea TM. Risk factors for organ

dysfunction and failure in patients with acute traumatic cervical spinal cord injury. Neurocrit Care 2010;13(1):29-39.

11

ACCEPTED MANUSCRIPT 13.

Byhahn C, Lischke V, Halbig S, Scheifler G, Westphal K. [Ciaglia blue rhino: a modified

technique for percutaneous dilatation tracheostomy. Technique and early clinical results]. Anaesthesist 2000;49(3):202-6. 14.

Ben Nun A, Orlovsky M, Best LA. Percutaneous tracheostomy in patients with cervical spine

fractures--feasible and safe. Interact Cardiovasc Thorac Surg 2006;5(4):427-9. Putensen C, Theuerkauf N, Guenther U, Vargas M, Pelosi P. Percutaneous and surgical

RI PT

15.

AC C

EP

TE D

M AN U

SC

tracheostomy in critically ill adult patients: a meta-analysis. Crit Care 2014;18(6):544.

12

ACCEPTED MANUSCRIPT Table 1. Patient characteristics at admission to the ICU

Table 2. Characteristics of each group at admission to the ICU Table 3. Treatment variables

AC C

EP

TE D

M AN U

SC

RI PT

Figure 1. Smoothed linear model

13

ACCEPTED MANUSCRIPT Table 1. Patient characteristics at admission to the ICU n (%)

Demographic variables

Female

8 (14.3)

Male

48 (85.7)

Age (years) Charlson’s comorbidity index ASIA classification

15.8 (15.0-80.0)

0.5±1.4

0.0 (0.0-8.0)

38 (67.9)

B

11 (19.6)

C

5 (8.9) 2 (3.6)

Motor index

15.8±15.5

12.0 (0.0-80.0)

Injury Severity Score (ISS)

26.8±12.2

25 (0.0-59.0)

SC

Lesion and severity variables Glasgow Coma Scale score at admission Total SOFA score at admission Total SOFA score on day 4 Type of surgery

13.6±3.1

15.0 (3.0-15.0)

11.6±6.7

10.0 (1.0-28.0)

M AN U

APACHE II score

Treatment variables

48.7±20.2

A

D

Anterior

5.4±3.0

5.0 (0.0-28.0)

3.7±2.1

4.0 (0.0-9.0)

54 (96.4)

Posterior

TE D

Days elapsed between the onset of the lesion and surgery Tracheostomy technique

16.4±14.3

Percutaneous

47 (83.9)

Surgical

9 (16.1)

2 (3.6)

14.0 (0.0-66.0)

14.6±10.6

12.5 (1.0-50.0)

Days of mechanical ventilation

66.2±65.2

41.0 (13.0-362.0)

AC C

EP

Days elapsed between the onset of the lesion and the tracheostomy procedure

Length of stay at the ICU (days) Length of the hospital stay (months) Death

Median (min-max)

RI PT

Sex

Mean±SD

34.0±15.2

30.0 (9.0-86.0)

6.9±4.4

7.0 (0.0-19.0)

Yes

10 (17.9)

No

46 (82.1)

ACCEPTED MANUSCRIPT Table 2. Characteristics of each group at admission to the ICU

NOT AT-RISK GROUP (N=25)

AT-RISK GROUP (N=31)

n (%)

n (%)

DEMOGRAPHIC VARIABLES

Female Male

Age (years) Charlson’s comorbidity index

1 (4.0)

7 (22.6)

24 (96.0)

24 (77.4)

Mean±SD

Median

Mean±SD

48.3±19.7

49.1

49.9±20.8

48.4

0.811

0.2±0.5

0

0.8±1.8

0

0.164

AT-RISK GROUP (N=31)

n (%)

n (%)

6 (24.0)

C5-D1

19 (76.0)

A

20 (80.0)

B

4 (16.0)

C

1 (4.0)

12 (38.7)

D

0.871

SC

C1-C4

21 (61.3)

18 (58.1) 7 (22.6)

M AN U

ASIA classification

Median

NOT AT-RISK GROUP (N=25) LESION AND SEVERITY VARIABLES

Lesion level

0.063

RI PT

Sex

P*

0 (0)

--

4 (12.9) 2 (6.5)

Mean±SD

Median

Mean±SD

Median

Motor index

13.5±8.5

12.0

17.8±19.6

12.0

0.985

Glasgow Coma Scale score at admission

13.6±2.8

15.0

13.5±3.4

15.0

0.882

Glasgow Coma Scale score at discharge from the ICU

15.0±0.0

15.0

14.5±2.1

15.0

0.177

Injury Severity Score (ISS)

26.3±11.4

27.1±13.0

25.0

0.825

10.0

11.7±7.6

9.5

0.754

Total SOFA score at admission

5.6±2.7

5.0

5.3±3.2

5.0

0.604

Total SOFA score on day 4

TE D

25.0

11.4±5.7

APACHE II score

3.0

4.2±2.4

4.0

0.138

1.8±1.3

2.0

1.8±1.2

2.0

0.945

Hemodynamic SOFA score on day 4

0.8±1.3

0.0

1.5±1.7

1.0

0.085

EP

3.1±1.6

Respiratory SOFA score on day 4

AC C

* The P-value is considered significant when P < 0.05

ACCEPTED MANUSCRIPT Table 3. Treatment variables NOT AT-RISK GROUP (N=25)

AT-RISK GROUP (N=31)

n (%)

n (%)

Yes

17 (68.0)

23 (74.2)

No

8 (32.0)

8 (25.8)

P*

Corticoids

0.610

Days elapsed between the onset of the lesion and surgery

Mean±SD

Median

Mean±SD

11.1±14.3

7.0

20.7±13.0

n (%)

18.0

<0.001

n (%)

Anterior

23 (92.0)

Posterior

2 (8.0)

36 (100.0)

Type of surgery

0.195

0 (0.0)

1

16 (64.0)

>1

9 (36.0)

19 (63.3)

Yes

3 (12.0)

No

22 (88.0)

Peri-paravertebral abscess

Yes No

TRACHEOSTOMY Days elapsed between the onset of the lesion and the tracheostomy procedure

Percutaneous

Tracheostomy technique

11 (36.7) 0 (0.0)

0.083

31 (100.0)

M AN U

Surgical wound infection

0.959

SC

Fixation Level

0 (0.0)

1 (3.2)

25 (100.0)

30 (97.8)

Mean±SD

Median

20.0±11.7

18.0

–-

Mean±SD

Median

10.2±7.3

8.0

n (%)

n (%)

19 (76.0)

28 (90.3)

0.001

TE D

0.272 Surgical

Stomal infection

6 (24.0)

3 (9.7)

Yes

0 (0.0)

0 (0.0)

No

25 (100)

31 (100)

Yes

4 (16.0)

3 (9.7)

No

21 (84.0)

28 (90.3)

EP

Non-infectious complications Decannulation (patients)

Median

RI PT

SURGERY

–-

0.688

Yes

15 (60.0)

18 (58.1)

No

10 (40)

13 (41.9)

0.884 Median

Mean±SD

Median

107.7±78.5

80.0

95.9±56.9

70.0

Mean±SD

Median

Days elapsed until initiation of oral nutrition

35.3±13.9

31.5

37.5±19.8

33.0

0.856

No. of days of sedation

10.5±7.1

11.0

14.4±10.4

11.0

0.283

No. of days of mechanical ventilation

66.4±61.9

45.0

66.0±68.7

40.0

0.947

Length of stay at the ICU (days)

35.2±19.4

28.0

32.9±11.0

30.0

0.843

Length of the hospital stay (months)

245.6±125.5

233.0

190.8±121.9

198.0

0.093

AC C

Mean±SD

Days elapsed between hospital admission and decannulation

n (%)

Mean±SD

0.481

Median

n (%)

Yes

2 (8.0)

8 (25.8)

No

23 (92.0)

23 (74.2)

Death

0.159

* the P-value is considered significant when P < 0.05

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

1.- Our results suggest that the presence of a tracheostomy stoma prior to, or immediately after surgery, is associated with a low risk of infection of the cervical surgical wound in instrumented spinal fusion. 2.- Adherence to good clinical practice guidelines during the surgical and perioperative

RI PT

periods can be a decisive factor in this respect.

AC C

EP

TE D

M AN U

SC

3.- The mortality rates of both groups (risk and not at-risk groups) did not differ significantly

ACCEPTED MANUSCRIPT

SCI: spinal cord injury ICU: intensive care unit MV: mechanical ventilation GCS: Glasgow Coma Scale ISS: Injury Severity Score

SOFA: Sequential Organ Failure Assessment

AC C

EP

TE D

M AN U

SC

SD: standard deviation

RI PT

APACHE II: Acute Physiology and Chronic Health Evaluation II