Accepted Manuscript Impact of perioperative fluid administration on postoperative morbidity and length of hospital stay following maxillomandibular advancement for obstructive sleep apnea Kyle S. Ettinger, DDS, MD, Cody C. Wyles, BS, Brett J. Bezak, DMD, Yavuz Yildirim, DDS, MD, Kevin Arce, DMD, MD, Christopher F. Viozzi, DDS, MD PII:

S0278-2391(15)00002-6

DOI:

10.1016/j.joms.2014.12.032

Reference:

YJOMS 56607

To appear in:

Journal of Oral and Maxillofacial Surgery

Received Date: 14 October 2014 Revised Date:

15 December 2014

Accepted Date: 23 December 2014

Please cite this article as: Ettinger KS, Wyles CC, Bezak BJ, Yildirim Y, Arce K, Viozzi CF, Impact of perioperative fluid administration on postoperative morbidity and length of hospital stay following maxillomandibular advancement for obstructive sleep apnea, Journal of Oral and Maxillofacial Surgery (2015), doi: 10.1016/j.joms.2014.12.032. 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: Impact of perioperative fluid administration on postoperative morbidity and length of hospital stay following maxillomandibular advancement for obstructive sleep apnea.

RI PT

Authors: Kyle S. Ettinger, DDS, MD*, Cody C. Wyles, BS**, Brett J. Bezak, DMD*, Yavuz Yildirim, DDS, MD*, Kevin Arce, DMD, MD***, Christopher F. Viozzi, DDS, MD****

Clinic and Mayo College of Medicine, Rochester, MN

SC

*Resident, Division of Oral and Maxillofacial Surgery, Department of Surgery, Mayo

M AN U

**Medical Student, Mayo Clinic and Mayo College of Medicine, Rochester, MN ***Assistant Professor of Surgery and Program Director, Division of Oral and Maxillofacial Surgery, Department of Surgery, Mayo Clinic and Mayo College of Medicine, Rochester, MN

TE D

****Assistant Professor of Surgery and Program Chair, Division of Oral and Maxillofacial Surgery, Department of Surgery, Mayo Clinic and Mayo College of Medicine, Rochester,

Correspondence

EP

MN

AC C

Kyle S. Ettinger, DDS, MD

Department of Surgery Mayo Clinic Rochester Division of Oral and Maxillofacial Surgery

Mail Code: ro_ma_12_12eres 200 First Street S.W Rochester, MN 55905 Email: [email protected]

ACCEPTED MANUSCRIPT

Abstract Purpose: The purpose of this study was to evaluate whether the volume of perioperative fluids administered to patients undergoing maxillomandibular advancement (MMA) for treatment of

complications and prolonged length of hospital stay.

RI PT

obstructive sleep apnea (OSA) is associated with increased incidence of postoperative

SC

Materials and Methods: A retrospective cohort study design was implemented and patients undergoing MMA for OSA at Mayo Clinic were identified from 2001-2013. The primary

M AN U

predictor variable was the total volume of intravenous fluids administered during MMA. The primary outcome variable was length of hospital stay in hours. Secondary outcome variables included the presence of complications incurred during postoperative hospitalization. Additional covariates abstracted included basic demographic data, preoperative body mass index,

TE D

preoperative apnea-hypopnea index, preoperative Charlson comorbidity index, preoperative American Society of Anesthesiologists score, the type of intravenous fluid administered, surgical complexity score, duration of anesthesia, duration of surgery, and the use of planned ICU

EP

admission. Univariate and multivariable models were developed to assess associations between the primary predictor variable and covariates relative to both primary and secondary outcome

AC C

variables.

Results: A total of 88 patients undergoing MMA for OSA were identified. Total fluid volume was significantly associated with increased length of stay (OR=1.34, 95% CI: 1.05-1.71, p=0.020) in univariate analysis. Total fluid volume did not remain significantly associated with increased length of hospital stay in stepwise multivariable modeling. Total fluid volume was

ACCEPTED MANUSCRIPT

significantly associated with the presence of postoperative complications (OR 1.69; 95% CI: 1.08-2.63; p=0.021) in univariate logistic regression.

RI PT

Conclusion: Fluid administration was not found to be significantly associated with increased length of hospital stay following MMA for OSA. Increased fluid administration is potentially associated with presence of postoperative complications following MMA; however, future large

SC

multicenter studies will be required to more comprehensively assess this association.

M AN U

Introduction

Since the initial characterization of obstructive sleep apnea (OSA) in 1976,1 healthcare providers across all medical and surgical specialties have become increasingly aware of the syndrome’s diagnostic clinical hallmarks and its pathophysiologic sequelae. Nevertheless, recent

TE D

epidemiologic studies estimate that as many as 1 in 20 adults are currently affected by OSA and many of these individuals are likely unrecognized and undiagnosed.2 Extrapolating to the population of the United States, as many as 15.8 million adult Americans are potentially affected

EP

by OSA. Coupled with increasing recognition of sleep disordered breathing in children,3-5 the number of individuals affected by OSA is unlikely to diminish in the foreseeable future. Given

AC C

this information, it is clear that untreated obstructive sleep apnea represents a significant public health burden; the annual cost associated with untreated OSA has been estimated to be $3.4 billion.6 Presently, positive airway pressure (PAP) therapy is the gold standard to which all other OSA treatments are measured. However, with as many as 50% of OSA patients ultimately intolerant of various forms of PAP therapy,7,8 alternative treatment modalities are of paramount clinical importance.

2

ACCEPTED MANUSCRIPT

Initially reported in the 1980’s,9-14 the use of maxillomandibular advancement (MMA) for the treatment of OSA has become an increasingly popular and successful strategy for managing patients intolerant of traditional PAP modalities. Maxillomandibular advancement has

RI PT

repeatedly been shown to have therapeutic efficacy equal to PAP therapy and it is now widely recognized to be the definitive surgical intervention for treatment of PAP intolerant patients when compared to alternative airway surgeries.15-17 However, despite a highly successful track

SC

record for improvement in polysomnographic indices such as apnea-hypopnea index (AHI), MMA is not without inherent risks and associated morbidity. Postoperative pharyngeal edema

M AN U

remains a real and omnipresent concern,18,19 which has led many to recommend the use of planned intensive care unit (ICU) admission as a means of more carefully monitoring and therefore possibly preventing potential airway embarrassment.20-25 Judicious administration of perioperative intravenous fluids has also been empirically recommended as a means of further

TE D

reducing postoperative laryngeal edema,24 yet at present, studies specifically evaluating the impact of fluid administration on patients undergoing MMA for OSA are nonexistent within the literature.

EP

Over the past decade there has been an increasing level of attention afforded to perioperative fluid management and its impact on postoperative surgical outcomes within the

AC C

field of anesthesiology.26-31 Liberal fluid administration has been correlated with impaired cardiac, pulmonary, and gastrointestinal function and also has been demonstrated to prolong postoperative surgical recovery.26,29,30 Conversely, restrictive fluid regimens can place patients at risk for postoperative hypovolemia, organ dysfunction, and in extreme cases multiple organ failure and death.32 However, in spite of numerous randomized clinical trials specifically evaluating the impact of perioperative fluid protocols on surgical outcomes, heterogeneity among

3

ACCEPTED MANUSCRIPT

studies and lack of standardization in fluid regimens has precluded definitive consensus regarding optimal fluid protocols to decrease postoperative morbidity and length of hospital stays.32 Nevertheless, optimization of perioperative fluid regimens is a readily modifiable aspect

RI PT

of surgical care that holds the potential for improving postoperative outcomes and shortening hospital stays—both of which are of critical importance for cost containment in healthcare

delivery. Given the already developing economic burden placed on the healthcare system by

SC

OSA and its associated sequelae, elucidation of the relationship between perioperative fluid

within the realm of OSA surgery.

M AN U

status and postoperative outcomes in MMA represents a potential area for quality improvement

The purpose of this study was to specifically evaluate impact of perioperative fluid administration on patients undergoing MMA for treatment of OSA. The investigators hypothesized that patients receiving higher perioperative fluid volumes would subsequently

TE D

experience greater third spacing of fluids, increased facial/pharyngeal edema, and consequently require longer hospitalizations with increased incidence of postoperative complications. The specific aims of the study were to measure and compare various fluid volumes administered to

EP

OSA patients undergoing MMA and determine whether the volume of fluids received correlated with an increased incidence of postoperative complications and prolonged length of hospital

AC C

stay. Further aims of the study also included the identification and description of additional factors associated with increased incidence of postoperative complications and prolonged hospital admission following MMA for OSA.

Materials and Methods Study Design/Sample

4

ACCEPTED MANUSCRIPT

To address the research purpose, the investigators designed and implemented a retrospective cohort study. A sample of subjects was derived from a population of patients being treated by the Division of Oral and Maxillofacial Surgery at Mayo Clinic in Rochester, MN

RI PT

between 2001 and 2014. Subjects were identified through an electronic database search of operative notes containing the keyword “sleep” within the preoperative or postoperative

diagnosis. Inclusion criteria for the study included maxillary and/or mandibular advancement

SC

(with or without advancement genioplasty) for the treatment of OSA within the specified time range, age greater than or equal to 12 years, adequate documentation of perioperative fluid

M AN U

administration, adequate documentation of postoperative hospital course and subsequent complications therein, and adequate documentation of dates of postoperative admission. Exclusion criteria for the study included independent or concomitant use of traditional Stage I surgical modalities33 for treatment of OSA (nasal septoplasty, turbinectomy,

TE D

uvulopalatopharyngoplasty, genioglossal advancement surgeries, isolated advancement genioplasty, hyoid myotomy/suspension, tongue volume reduction surgery), age less than 12 years, inadequate documentation of intraoperative fluid administration, inadequate

EP

documentation of postoperative hospital stay and associated complications, lack of documentation of dates of postoperative admission, autogenous iliac crest bone grafting as a

AC C

component of maxillomandibular advancement surgery, history of cleft lip and/or palate, any concomitant craniofacial syndrome, performance of simultaneous ancillary procedures in addition to standard maxillomandibular advancement surgery (i.e. cervicofacial liposuction, bilateral or unilateral prosthetic TMJ replacement, etc.), and surgery performed as a reoperation for previously failed maxillomandibular advancement. Approval from the Mayo Clinic Institutional Review Board was obtained for completion of this study.

5

ACCEPTED MANUSCRIPT

Variables The primary predictor variable for the study was the total volume of perioperative fluids

RI PT

administered on the day of surgery. The total volume of perioperative fluids administered

represents the combination of both colloid and crystalloid fluids volumes administered to the patient preoperatively in the preop holding area, intraoperatively during the procedure, and

SC

postoperatively until midnight on the day of the surgery. This method of calculating

perioperative fluid volume was selected in order to standardize the recording of fluid data among

M AN U

the subject population and to facilitate ease of abstraction from the electronic medical record. The primary outcome variable for the study was the length of hospital stay in hours following completion of the MMA for OSA. The length of hospital stay was recorded as the time elapsed in hours between the patient being admitted to the receiving ICU or general hospital

TE D

floor following their procedure to the exact time of patient dismissal from the hospital as recorded in the electronic medical record. The method of calculating length of hospital stay was intentionally chosen to eliminate the time of the surgical procedure and is consequently why the

EP

start of admission occurs upon the patient reaching either the ICU or general hospital floor following the procedure.

AC C

Secondary outcome variables for the study included the presence of various

complications incurred during the postoperative hospitalization. These complications were intentionally chosen (with minor additions and modifications) to mirror those set for by the American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP), which is the most widely recognized quality measurement system for non-cardiac surgery in the United States.34 The categories of postoperative complications and their definitions are as

6

ACCEPTED MANUSCRIPT

follows: 1) pulmonary (any unplanned intubation/mechanical ventilation following normal postoperative extubation, mechanical ventilation >48 hours, pneumonia, continued postoperative intubation beyond the PACU for any duration); 2) cardiovascular (ST elevation myocardial

RI PT

infarction, non-ST elevation myocardial infarction, heart failure, cardiac arrest, any new cardiac arrhythmia requiring medical treatment); 3) renal (acute kidney injury representing twofold increase in serum creatinine or urine output < 0.5 mL/kg/hr for 12 hours, acute renal failure

SC

representing threefold increase in serum creatinine or urine output < 0.3 mL/kg/hr for 24 hours or anuria for 12 hours); 4) genitourinary (urinary tract infection); 5) neurologic (ischemic or

M AN U

hemorrhagic cerebrovascular accident, coma); 6) hematologic (deep vein thrombosis, pulmonary embolism); 7) infectious (sepsis/septic shock); 8) postoperative blood transfusion; 9) unscheduled ICU admission. Postoperative complications occurring beyond the initial inpatient hospitalization were not recorded.

TE D

Multiple covariates were also abstracted for the study and included: patient age (in years); preoperative body mass index (BMI); preoperative apnea-hypopnea index (AHI); duration of anesthesia (in hours); duration of surgery (in hours); sex (male/female); smoking

EP

status (never smoked, active smoker, former smoker); Charlson comorbidity index;35 preoperative American Society of Anesthesiologists score (ASA score); complexity of surgical

AC C

procedure; specific type of colloid fluid administered (albumin, hydroxyethyl starches, packed red blood cells); and presence of a planned ICU admission (defined as the preoperative decision to admit the patient to an ICU setting following MMA irrespective of the postoperative outcome). Duration of anesthesia was recorded as the time elapsed in hours between the anesthesiology team assuming care of the patient upon entrance into the OR to the exact time of the post-anesthesia care unit (PACU) assuming care of the patient upon completion of the

7

ACCEPTED MANUSCRIPT

procedure. Duration of surgery was recorded as the time in hours from the documented time of incision to the documented time of closure within the electronic anesthesia record. The purpose of distinguishing between duration of anesthesia and duration of surgery was intentionally

RI PT

chosen in order to stratify differences among outcomes specifically resulting from anesthesia delays (i.e. delays in obtaining vascular access, delays in induction of general anesthesia, delays in postoperative extubation, etc) relative to intraoperative surgical delays resulting from

SC

challenges within the procedure itself. The Charlson comorbidity index35 represents a validated scoring system that predicts the ten-year mortality for a patient based on the presence of 22

M AN U

specific comorbid disease conditions. Each comorbid condition carries a specific score weighting based on the risk of mortality associated with the disease. The weighting system is as follows: 1 point (myocardial infarction, congestive heart failure, peripheral vascular disease, dementia, cerebrovascular disease, chronic lung disease, connective tissue disease, chronic liver

TE D

disease, diabetes); 2 points (hemiplegia, moderate or severe kidney disease, diabetes with end organ damage, any tumor, leukemia, lymphoma); 3 points (moderate or severe liver disease); 6 points (malignant tumor, metastatic disease, AIDS). The summation of point values for each

EP

condition represents the overall comorbidity score. The complexity of the surgical procedure was quantified through the use of a previously described surgical complexity score.36 The type

AC C

of maxillary procedure was recorded as a 1-piece LeFort, 2-piece LeFort, 3-piece LeFort, or not performed. The type of mandibular procedure was recorded as a bilateral sagittal split ramus osteotomy or not performed. The type of chin procedure was recorded as a 1 piece genioplasty or not performed. The complexity score for the procedure was set to an initial baseline value of 0 with 1 point added for a 1-piece LeFort, 1.5 points for a 2-piece LeFort, 1.75 points for a 3piece LeFort, 1 point for a bilateral sagittal split ramus osteotomy, and 1 point for advancement

8

ACCEPTED MANUSCRIPT

genioplasty. A summative value of the individual procedures performed during the operation was then calculated to represent the overall surgical complexity score. Perioperative steroid administration was not specifically abstracted as a covariate within

RI PT

the study. All subjects undergoing MMA within the specified time range were provided a single dose of intravenous steroids at the time of the procedure in the form of an 8- or 10-mg dose of intravenous dexamethasone depending on surgeon preference. Postoperatively all patients were

SC

placed on a scheduled steroid regimen consisting of intravenous dexamethasone every 12 hours for a total of 5 doses. Patients were not required to complete all 5 doses prior to dismissal. If

M AN U

less than 5 doses of steroids were administered, no additional outpatient steroids were provided. This steroid protocol remained constant throughout the years in which the subject pool was garnered (2001-2014).

TE D

Data Collection

The electronic medical records of the subjects in the study pool were retrospectively reviewed for abstraction of all necessary study variables and demographic data. The collected

EP

data was securely stored in a Mayo Clinic sponsored Research Electronic Data Capture (REDCap) database designed specifically for the research project by the primary author (KE).

AC C

All authors participating in data abstraction were calibrated on the appropriate use and population of the REDCap database prior to initiating chart review. Access to the REDCap database was limited to the primary author, the authors involved with data abstraction, and the project biostatistician. Any missing data following initial abstraction attempts was subsequently identified and incorporated into the database by the primary author. Data Analysis

9

ACCEPTED MANUSCRIPT

Total fluid was calculated as the sum of crystalloid and colloid volumes administered perioperatively. Length of stay was grouped as

Impact of Perioperative Fluid Administration on Postoperative Morbidity and Length of Hospital Stay Following Maxillomandibular Advancement for Obstructive Sleep Apnea.

The purpose of this study was to evaluate whether the volume of perioperative fluids administered to patients undergoing maxillomandibular advancement...
309KB Sizes 0 Downloads 8 Views