Design and Impact of Intraoperative Pathways for Head and Neck Resection and Reconstruction

Objectives  To describe the design and impact of 3 intraoperative pathways for the treatment of head and neck cancers; to detail the pathways schematically to illustrate projected intraoperative flow and teamwork; and to analyze impact on procedure and case lengths in each pathway and in comparison with historical prepathway average times.

Setting  Tertiary-level academic health system main operating room.

Patients  Twenty-one patients undergoing transcervical (TC) resection (n = 11), transmandibular (TM) resection (n = 8), or laryngopharyngectomy (LP) (n = 2) with radial forearm free-flap reconstruction for ablative or reconstructive reasons were pathway eligible. A convenience sample of 16 patients undergoing TC resection, 7 undergoing TM resection, and 7 undergoing LP prepathway is used for comparison.

Intervention  Our academic medical center uses 3 intraoperative clinical pathways to manage resource use and streamline care for patients. These 3 pathways were designed schematically by an interdisciplinary team. The pathways plan progression of the case by timed actions for surgical, anesthesia, and nursing teams.

Main Outcome Measures  Procedure and case lengths.

Results  The TC pathway procedure and case length averaged 10.48 and 12.33 hours, respectively; TM pathway procedure and case lengths, 11.19 and 13.32 hours, respectively; and LP pathway procedure and case lengths, 12.42 and 13.83 hours, respectively. Aggregate averages were 10.93 hours and 12.85 hours for procedure and case length, respectively. The average pathway case lengths of 12.33, 13.32, and 13.83 hours compare favorably with our target times of 13, 14, and 15 hours, respectively. Environmental management, work flow, and team satisfaction anecdotally increased postpathway.

Conclusions  Intraoperative pathways afford enhanced time and action efficiency to streamline care of patients undergoing head and neck procedures. Pathway implementation produced time savings. Our results suggest that implementation of such pathways will benefit similar academic medical centers seeking to improve intraoperative resource use to improve performance in the care of patients undergoing head and neck procedures.

WE DESCRIBE the design and clinical impact of 3 intraoperative pathways: transcervical (TC) resection of oral cavity neoplasia, transmandibular (TM) resection of oral or oropharyngeal neoplasia, and laryngopharyngectomy (LP), all with radial forearm free-flap (RFFF) reconstruction. This article details the design of each pathway that projects intraoperative flow and teamwork. The impact of the pathways is analyzed through descriptive statistics for procedure and case lengths in each pathway, against target pathway times, and in inferential comparison with grouped prepathway times. Summary conclusions include time and action efficiencies, projections for further analysis, and directions for research and practice.

Our academic medical center uses 3 intraoperative clinical pathways to manage resource use and streamline care for patients undergoing head and neck resections with RFFF reconstruction. These intraoperative pathways were developed in parallel with a head and neck clinical management pathway. The results of this pathway are reported elsewhere.¹ The intraoperative pathway was separated to enable greater specificity in work and resource redesign to improve patient care.

These 3 intraoperative pathways were the first for the Department of Otorhinolaryngology/Head and Neck Surgery, Hospital of the University of Pennsylvania. Our institution has been performing microvascular reconstruction in patients undergoing head and neck surgical procedures for 10 years. Head and neck resection with reconstruction has been viewed as difficult to standardize because of extended-length cases with multiple components and individual anatomic and design elements. The TC resection of oral cancers, the TM resection of oral and oropharyngeal cancers, and LPs, all with RFFF reconstruction, were selected for pathways because of the high positive prepathway profile of these operations and because of the head and neck team members (A.A.C., A.N.G., A.D., G.S.W., and R.S.W.) within the intraoperative environment.

The goal of the design was to create an algorithm that would specify a time-action pathway with time- and discipline-specific resources and actions to maintain patient outcomes and potentially achieve improved outcomes by shortening case length. Improved patient outcomes were defined as decreased time in the operating room (OR) and decreased anesthesia time. Cost savings and enhancedclinical and nonclinical staff satisfaction were assumed to be imbedded in achieving patient outcomes.

The 3 intraoperative head and neck surgery pathways were designed schematically by an interdisciplinary team (A.A.C., A.N.G., A.D., G.S.W., and R.S.W.). They were part of a larger institutional initiative that had 5 goals: specialty-focused intraoperative pathway development; reduce unnecessary variation and use in materials, durable equipment, medications, staff, and time; increase accuracy, quality, completeness, and timeliness of all resources used; increase direct patient care for nursing staff; and increase throughput to decrease cost per case.

The primary attending head and neck reconstructive surgeon (A.A.C.), the attending anesthesiologist (A.N.G.) for the OR section, and the primary circulating nurse (A.D.) for patients undergoing head and neck procedures formed the core team. Attending head and neck surgeons (A.A.C., G.S.W., and R.S.W.), attending anesthesiologists (A.N.G. and others), perioperative nurses (A.D. and others), and perioperative administrators reviewed the pathways during development, contributing to the design as necessary. The specific goals of the head and neck core team were to achieve standardized timed actions for surgery, anesthesiology, and nursing; to standardize the instrument sets for each pathway; to create a routine environmental setup; and to accommodate the individual components of each patient's operation (eg, the extent of the tumor; anatomic variation, such as neck exposure; and unilateral vs bilateral neck dissection). The pathway goals assumed synchronous flap harvest. This assumption added the potential variation of team surgery coordination.

The pathways plan progression of the case by timed actions for surgical, anesthesia, and nursing teams. The team member responsibilities were detailed in schematic form for each pathway. A distillation of these roles and responsibilities is presented in Table 1. The pathway is predicated on a single nursing team with 2 staff scrubbed and 2 surgical teams as necessary to resect the tumor and reconstruct the defect. This team structure was in place before pathway implementation and was systematized by the pathway.

A basic pathway schema was developed for consistency (Figure 1). Case length was defined as time into room to time out of room. Procedure length was defined as time from skin incision to time dressings were placed on the patient. The intraoperative environment was examined and resource needs delineated. Instrument set development was a major focus for resource use and clinician satisfaction. Eligibility criteria included neoplasia of the oral cavity, hypopharynx, oropharynx, or soft tissues of the head and neck. Tumor stage and radiation status were not part of the eligibility criteria. Patients were ineligible if they had skull base tumors; bone, pectoralis, or latissimus dorsi flaps; or lateral thigh or rectus abdominus free flaps. The default criterion for the pathways was premature termination or a canceled procedure.

The pathway schematics are formatted left to right, starting at time minus 30 minutes, with OR setup and temperature controlled from the night before. Team responsibilities are described along the horizontal axis for the surgeon, the anesthesiologist, and the nurse. These times are sequenced by actions across the top of the pathway schematic to summarize the actions taken by team members that are listed on the vertical axis of the actual pathway schematic (Figure 1). The pathway is completed at the point the room is readied for the next patient. Target pathway times were determinedby consensus because no benchmarking data were available. Case length was estimated from the OR's surgical database for historically derived averages. The historical time for TC resection was 16 hours; TM resection, 17 hours; and LP, 19 hours. These times were used primarily for design purposes. The target pathway times for case length, developed by the pathway team, are 13 hours for TC resection, 14 hours for TM resection, and 15 hours for LP.

Pathway implementation occurred in March 1998. Twenty-one patients undergoing reconstructive surgery, for the first year after implementation, were enrolled in the pathways for TC resection (n = 11), TM resection (n = 8), or LP (n = 2), with RFFF reconstruction for all. Data for 4 patients were incompletely recorded, and could not be used in analysis. These patients are analyzed using descriptive statistics and qualitative variables for work flow and team satisfaction. Prepathway averages from a convenience sample of 30 surgical procedures (16 TC resections, 7 TM resections, and 7 LPs) performed in the 2 years preceding pathway implementation are used for comparison. Prepathway and postpathway surgical procedures were performed by the same team of surgeons (A.A.C., G.S.W., and R.S.W.), using the same simultaneous flap harvest techniques.

Three times are used for comparison: ischemia time during microvascular reconstruction, procedure length, and case length. Ischemia time was recorded as the time (in minutes) for action 2 on the pathway (Figure 1). Procedure length was recorded in minutes from skin incision to dressing. Pathway case length, as noted earlier, was defined as time into the OR to time out of the OR, inclusive and recorded in minutes. All recorded times were converted to hours.

The targeted goals of decreased procedure time, representing anesthesia time, and decreased overall case length were achieved (Table 2). Aggregate pathway mean times (N = 21) were 10.93 hours for procedure length and 12.85 hours for case length. The aggregate postpathway case lengths are statistically significantly different from the prepathway case lengths. The pathway mean case length also achieved the target time for the TC pathway (mean, 12.85 hours; SD, 1.76 hours; range, 10.50-16.60 hours), undercutting the target times for the TM and LP pathways. The mean time for procedure length was 10.93 hours (SD, 2.18 hours; range, 5.80-15.57 hours); and for ischemia, 3.28 hours (SD, 0.77 hours; range, 1.56-4.67 hours).

The aggregate postpathway procedure lengths are statistically significantly different from the prepathway procedure lengths (mean, 10.93 vs 12.50 hours; t test; P = .009). The postpathway ischemia times are not statistically significantly different from the prepathway ischemia times at P = .05 (mean, 3.28 vs 3.58 hours; t test; P = .07). The mean ischemia times did, however, show a clinical trend toward a decline of 0.3 hours postpathway.

Transcervical pathway procedure and case lengths averaged 10.48 and 12.33 hours, respectively. This mean case length was shorter than the target of 13 hours. In addition, the procedure length was statistically different from the prepathway procedure length (mean, 10.48 vs 12.37 hours; t test; P = .02). The case length was also statistically different from the prepathway case length (mean, 12.33 vs 14.45 hours; t test; P = .009).

The TM and LP pathways had limited enrollment. Prepathway and postpathway analyses were not statistically valid or significant because of the small subsample sizes. Transmandibular pathway procedure and case lengths averaged 11.19 and 13.32 hours, respectively. This is shorter than the target of 14 hours. Laryngopharyngectomy pathway procedure and case lengths averaged 12.42 and 13.83 hours, respectively. This is considerably shorter than the target of 16 hours.

Environmental management, work flow, and team satisfaction were anecdotally noted to increase postpathway implementation. Inconsistent instrument availability was a prepathway dissatisfier for surgeons and nurses. The instrument set was standardized through the pathway design and implementation. An instrument processing partner was assigned for the pathways. Pathway implementation leads to significant improvement in appropriate instrument availability. No standardized measures of work flow or satisfaction were incorporated into the project design. However, team members believed that the design process and implementation strengthened their collaboration.

Organizational analysis and redesign of the health care industry is an omnipresent topic in clinical literature without rigorous traditional research to support specific uses.² Many institutions have redesigned work groups and processes, based largely on organizationally perceived cost-benefit analyses, with and without the assistance of external consultants. Historically, processes with little tolerance for error were schematized (eg, advanced cardiac life support protocols). Resource and time-intensive operative and other invasive procedures requiring anesthesia frequently lend themselves to redesign. Goals most often include standardized resources, such as instruments, and schematized time and action algorithms. Processes selected for redesign are generally high volume, high resource, or high cost within an organization or subspecialty.

Perioperative and invasive procedure redesign initiatives reported in medical and nursing literature include the following: frontal sinus obliteration,³ head and neck oncologic surgery and chemotherapy,⁴ ambulatory surgery,⁵ carotid endarterectomy,⁶ radical prostatectomy,⁷ knee replacement,⁸ and elective colon resection.⁹ The reports consistently refer to goals of improved resource use without adverse effects to patients for procedures that were clinically predictable intraoperatively and perioperatively. Reported pathways excluded intraoperative processes,^3,6,7 redesigned anesthesia intraoperative process alone or only,⁵ or redesigned intraoperative and perioperative processes.⁴ All researchers reviewed achieved goals of improved resource use while maintaining quality of care, as measured in each project, through systematizing their selected operations or procedures.

Correa and colleagues³ analyzed a pathway for frontal sinus obliteration. The design process detailed is similar to that used in this project. Intraoperative and postoperative processes and outcomes were included. The project sought to decrease operative time and estimated blood loss. These outcomes were achieved. Cohen and colleagues⁴ report on the development and implementation of pathways designed to reorganize head and neck oncologic care, focusing on chemotherapy and postsurgical care. While statistically significant improvements in length of hospital stay and cost of hospitalization were achieved, intraoperative processes were not a focus of the project. The present project, when compared across specialties, is most similar to the description of Baker and colleagues⁵ of redesigning ambulatory surgical procedures. Baker and colleagues schematized intraoperative time-action algorithms for their institution's same-day surgery unit, achieving economies of scale in resource use and improved postoperative patient and regulatory compliance.

Our experience supports implementation of similar pathways in a comparable large academic medical center, with a significant head and neck surgery patient volume, seeking to improve intraoperative resource use and to garner enhanced performance in the care of these patients. Our intraoperative pathways afford enhanced time and action efficiency to streamline the care of patients undergoing head and neck procedures. The mean aggregate case length achieved the target time for the TC pathway. This was the most limited of the 3 procedure target goals. This suggests that the process of intraoperative pathway implementation is valuable and that performance improvement in time-action efficiencies can be achieved. It also suggests that the accommodation for procedure may not be necessary if approaches to resection and to reconstruction are appropriately matched. Our experience supports the notion that complex operations with multiple components may be standardized. We acknowledge the need for allowances that reflect tumor extent, anatomic variation that may limit the efficiency of resection (eg, neck exposure, obesity, or another comorbidity), possible metastatic disease that requires bilateral neck dissection, and free-flap design elements.

Further analysis of cost benefits is pending acquisition of information processing systems for cost per OR hour and cost per case. Comparative analysis is limited by the nature of available literature and by the extent of our database. Future research that builds information management systems and enables detailed financial analysis within the context of hospital stay and billing returns will add an additional dimension to this area of outcomes research and performance improvement.

Accepted for publication January 9, 2002.

We thank the Head and Neck Intraoperative Pathway Team, Hospital of the University of Pennsylvania, Philadelphia.

Corresponding author: Ara A. Chalian, MD, Department of Otorhinolaryngology/Head and Neck Surgery, Hospital of the University of Pennsylvania, 5 Silverstein, 3400 Spruce St, Philadelphia, PA 19104-4283 (e-mail: chaliana@uphs.upenn.edu).