Postoperative Complications of Paramedian Forehead Flap Reconstruction

This cohort study analyzes records of 2175 patients who underwent paramedian forehead flap reconstruction to evaluate the occurrence of postoperative complications.

Key Points

Question

What are the rates of postoperative complications and hospital readmissions in patients receiving paramedian forehead flap surgery?

Findings

In this cohort study of data from 2175 patients, overall complication rates were low, including postoperative deep venous thrombosis. The most common complication was postoperative infection, while postoperative bleeding had the greatest odds of hospital readmission within the first 48 hours of surgery.

Meaning

Forehead flap surgery can be safely performed for patients with facial cancer in the outpatient setting without the need for anticoagulation; postoperative bleeding in this population carries the greatest risk for readmission.

Abstract

Importance

Paramedian forehead flaps are commonly used to reconstruct facial defects caused by skin cancers. Data are lacking on the complications from this procedure, postoperative outcomes, and association of cancer diagnosis with rate of deep venous thrombosis (DVT).

Objectives

The primary objective was to determine complication rates after paramedian forehead flap reconstruction for defects resulting from resection of facial cancers; and the secondary objective was to determine patient factors and complications that are associated with readmission.

Design, Setting, and Participants

Retrospective cohort study of patients who underwent paramedian forehead flap reconstruction for skin cancer reconstruction from January 1, 2007, through December 31, 2013. Data analysis took place between October 1, 2017, and June 1, 2018.

Main Outcomes and Measures

Complication rates including DVT, emergency department visits, and hospital readmissions.

Results

A total of 2175 patient were included in this study; mean (SD) age, 70.3 (13.4) years; 1153 (53.5%) were men. Postoperative DVT occurred in 10 or fewer patients (≤0.5%); postoperative bleeding in 30 (1.4%), and postoperative infection in 63 (2.9%). Most patients went home on the day of surgery (89.6%; n = 1949), while 10.4% stayed one or more days in the hospital (n = 226). Overnight admission was associated with tobacco use (odds ratio [OR], 1.65; 95% CI, 1.11-2.44), hypothyroidism (OR, 1.93; 95% CI, 1.10-3.39), hypertension (OR, 1.82; 95% CI, 1.29-2.57), ear cartilage graft (OR, 2.20; 95% CI, 1.51-3.21), and adjacent tissue transfer (OR, 1.88; 95% CI, 1.33-2.67). Risk factors strongly associated with immediate return to the emergency department or readmission within 48 hours of surgery included postoperative bleeding (OR, 13.05; 95% CI, 4.24-40.16), neurologic disorder (OR, 4.11; 95% CI, 1.12-15.09), and alcohol use (OR, 7.70; 95% CI, 1.55-38.21).

Conclusions and Relevance

In this study, the most common complication of paramedian forehead flap reconstruction was infection. Risk factors for readmission included development of postoperative bleeding, having a neurologic disorder, and alcohol use. Deep venous thrombosis was a rare complication. Because bleeding is a more common complication in this patient population, discretion should be used when deciding to administer anticoagulation medication to low- to medium-risk patients prior to surgery.

Level of Evidence

NA.

Introduction

The face is particularly vulnerable to cutaneous cancers, both melanotic and nonmelanotic. With advances in Mohs surgery, resection of malignant cutaneous facial lesions can be accomplished with more reliable margins, leading to improved cure rates while preserving surrounding tissue.^1,2 These advances have allowed for safe and effective resection of cutaneous cancer without leaving behind unmanageably large defects, which previously resulted from wide local excision.³

While the accessibility of Mohs surgery has transformed the treatment of facial cutaneous cancer, it has also transformed the surgical treatment of the facial defects left behind. Although a great number of resection defects can be reconstructed with local advancement flaps, rotational flaps, and transposition flaps, larger defects require reconstruction with substantial tissue bulk supported by an axial blood supply.⁴ Surgeons have relied on the paramedian forehead flap for decades owing to its robust blood supply from the supratrochlear artery and its versatility when combined with additional tissue types used during reconstruction, such as cartilage and bone grafts.^5,6,7,8

The paramedian forehead flap is usually performed as a 2- or 3- stage procedure, and in certain situations it can be performed as a single-stage surgery.⁹ As a result, most patients receiving forehead flaps will need at least 1 additional procedure to divide the vascular pedicle, and potentially more surgeries to refine the site of inset with debridement, skin grafting, dermabrasion, and other procedures. Furthermore, because the forehead flap is an interpolated flap requiring careful dissection and complicated inset, patients may experience extensive operating times.

The postoperative complications of paramedian forehead flap surgery have been documented in single-institution settings through retrospective reviews. Previous studies have also investigated preoperative wound factors and patient factors associated with postoperative complications.^10,11,12 New vascular imaging technologies have resulted in decreased time between stages and theoretically reduce the risk of flap necrosis.^13,14 However, to our knowledge, there are no large studies that track patients during the postoperative period to examine medical and surgical management of their complications. Additionally, data are lacking on the rate of postoperative venous thromboembolism (VTE) as a specific complication after this extensive procedure.

The primary goal of this study was to determine the complication rate in a large cohort of patients undergoing paramedian forehead flap surgery over a sufficiently long follow-up period to identify important outcomes. Specifically, we analyzed the rate of VTE in the postoperative period in addition to other surgical complications. The secondary goal was to determine the postoperative management course of forehead flap patients. Specifically, we analyzed patient and procedural factors associated with scheduled admission after surgery as well as the postoperative rates of unplanned readmissions and emergency department (ED) visits within the first 48 hours.

Methods

Study Design

This secondary data analysis reviews the records of a cohort of patients undergoing paramedian forehead flap reconstruction with concurrent skin cancer diagnoses between January 1, 2007, and December 31, 2013, in California, Florida, and New York. All data analysis took place between October 1, 2017, and June 1, 2018. The institutional review boards of Washington University School of Medicine and Stanford University School of Medicine deemed this study exempt from oversight and patient written informed consent for deidentified data by federal definitions of their jurisdiction.

Data Sources

The Healthcare Cost and Utilization Project (HCUP) comprises a group of health care databases, related software tools, and products sponsored by the Agency for Healthcare Research and Quality (AHRQ). The present study uses 3 HCUP databases: the State Ambulatory Surgery Databases (SASD),¹⁵ the State Inpatient Databases (SID),¹⁶ and the State Emergency Department Database (SEDD)¹⁷ from California, Florida, and New York. These statewide databases contain information from discharge records for all patients, regardless of age (pediatric and adult) or payer (Medicare, Medicaid, private insurance, and no insurance). The SASD contains ambulatory surgery records from hospitals or free-standing ambulatory surgical centers; the SID contains records from inpatient hospital visits; the SEDD contains records from ED visits.

Individual patients are linked and tracked across all 3 databases with an encrypted patient-level identifier. An encrypted variable for admission date and length of stay was used to calculate the time period between visits for each patient, while the exact dates were kept encrypted to protect patient confidentiality.¹⁶

Study Population

Patients who underwent paramedian forehead flap surgery with a concurrent skin cancer diagnosis, melanotic and nonmelanotic, in an outpatient surgery center were included in the study. The study contains data from January 1, 2007, to December 31, 2013, from the states of Florida and New York, and from January 1, 2007, to December 31, 2011, from the state of California. Patients were identified in the SASD using Current Procedure and Terminology (CPT) codes for paramedian forehead flap (15731) and International Classification of Diseases, Ninth Revision (ICD-9) codes for skin cancer diagnoses (172.x, and 173.x). A 1-year follow-up period was required, and patients were required to be residents of the state in which the procedure was performed to limit loss to follow-up. Identified cases in the SASD were then linked by encrypted patient identifier to hospitalizations in the SID and ED encounters in the SEDD to obtain information regarding complications contained within these databases.

Data Collection

Data collected included patient demographic variables at the time of surgery. Age was divided into 4 groups (<40, 41-60, 61-75, and >75 years). Race was categorized as white, black, Hispanic, or other. Primary payer was categorized as Medicare, Medicaid, private insurance, self-pay, no charge, or other. Additional procedures performed concurrently with forehead flap surgery were also tracked, such as ear cartilage graft and adjacent tissue transfer. eTable 1 in the Supplement provides a complete list of codes used in the database query. Postoperative admission status was categorized as no admission, admission for 1 day, and admission for 2 to 4 days.

The primary outcome measures were complications from forehead flap surgery, including DVT, bleeding, infection, and sepsis during the 1-year follow-up time. Additionally, this study tracked rates of postoperative ED visits and readmissions within 48 hours and within 30 days of surgery. Specific comorbidities were examined to determine their possible associations with postoperative admission, ED visits, and immediate readmissions. Most patient comorbidities were defined using the comorbidity measure detailed by Elixhauser et al.¹⁸ This measure contains a group of 30 comorbidities using ICD-9 codes found to be significantly associated with in-hospital mortality. Patient comorbidities and corresponding diagnosis codes assessed in this study are also detailed in eTable 1 in the Supplement.

Statistical Analysis

Standard descriptive statistics were used to describe the study population, including demographic data, patient diagnoses and comorbidities, procedural characteristics, length of hospital stay, timing of revisits, and rates of complications. The HCUP Data Users Agreement does not permit disclosure of any findings that may be used to identify unique individuals in the data. This includes reporting subcategories that include fewer than 11 patients. For this reason, rare events captured in our database analysis are reported as less than or equal to 10 (≤10). Univariable logistic regression was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for factors related to both postoperative admissions and hospital or ED readmissions on postoperative day 0 or 1. All patient characteristics reaching statistical significance in the univariable model were included in multivariable logistic regression. Diagnostic tests, including tests of collinearity, were used to assure that all assumptions of the final model were met. Software programs (SAS 9.3 and SAS Enterprise Guide; both from SAS Institute Inc) were used for all database management and statistical analyses.

Results

A total of 2175 patients were included in this study. The most common diagnosis code was for malignant neoplasm of the face (ICD-9 173.3, 39% of patients; n = 848), and a specific diagnosis of basal cell carcinoma (ICD-9 173.31, 15.8% ; n = 344) was 2.4 times more common than for squamous cell carcinoma (ICD-9 173.32, 6.7%; n = 146). A small percentage of patients (4.0%; n = 87) had a facial melanoma. The majority of patient data came from the state of Florida (67.3%; n = 1612). The mean (SD) age of the patient cohort was 70.3 (13.4), years and 53.5 % of patients were male (n = 1153). Medicare was the primary payer (67.2%; n = 1461). The most common procedure performed concurrently with forehead flap was adjacent tissue transfer (20.6%; n = 448), followed by ear cartilage graft (13.1%; n = 286), and skin grafting (8.8%; n = 191). Most patients went home on the day of surgery (89.7%; n = 1716), while 10.4% of patients; n = 202) stayed 1 or more days in the hospital (Table 1).

Table 1. Patient and Procedure Characteristics.

Characteristic	Patients, No. (%)
Total	2175 (100)
Sexa
Male	1153 (53.5)
Female	1004 (45.5)
State
California	411 (17.2)
Florida	1612 (67.3)
New York	371 (15.5)
Age, mean (SD), y	70.3 (13.4)
Raceb
White	1948 (93.1)
Black	≤10 (≤0.5)
Hispanic	80 (3.8)
Other	60 (2.8)
Primary payer
Medicare	1461 (67.2)
Medicaid	48 (2.2)
Private insurance	565 (26.0)
Self-pay	32 (1.5)
No charge	16 (0.7)
Other	53 (2.4)
Diagnosis
Malignant neoplasm face	848 (39.0)
Basal cell carcinoma	344 (15.8)
Squamous cell carcinoma	146 (6.7)
Melanoma	87 (4.0)
Malignant neoplasm, site unspecified	750 (34.5)
Additional procedures
Complex wound repair	78 (3.6)
Split-thickness skin graft	30 (1.4)
Full-thickness skin graft	161 (7.4)
Adjacent tissue transfer	448 (20.6)
Myocutaneous flap	67 (3.1)
Ear cartilage graft	286 (13.1)
Composite graft	62 (2.9)
Rib graft	≤10 (≤0.5)
Admissionc
None	1716 (89.7)
1 Day	181 (9.5)
2-4 Days	18 (0.8)

^a18 Missing entries.

^b82 Missing entries.

^c261 Missing entries.

The rate of postoperative DVT was low, occurring at a rate of 0.5% or lower (≤10 of 2175). Infection was the most commonly coded postoperative complication, occurring at a rate of 2.9% (63 of 2175), with a 1.0% rate of sepsis (21 of 2175) and a 0.8% rate of intensive care unit admission (18 of 2175). The rate of postoperative bleeding was 1.4% (30 of 2175). Eighty-six patients (4.0% of total) returned to the ED within 30 days of surgery, and 31 (1.4% of total) returned within the first 48 hours of surgery. Fifty-four patients (2.5% of total) were readmitted to the hospital within 30 days of surgery, with 10 or fewer readmissions occurring within the first 48 hours of surgery (≤0.5% of total) (Table 2).

Table 2. Postoperative Complications.

Complication	Patients, No. (%)
Deep vein thrombosis	≤10 (≤0.5)
Bleeding	30 (1.4)
Infection	63 (2.9)
Sepsis	21 (1.0)
Critical care admission	24 (1.1)
Emergency department visit
Within 48 h	31 (1.4)
Within 30 d	86 (4.0)
Readmission
Within 48 h	≤10 (≤0.5)
Within 30 d	54 (2.5)

After multivariable logistic regression, factors found to have a statistically significant association with postoperative admission were concurrent adjacent tissue transfer (OR, 1.88; 95% CI, 1.33-2.67; P < .001) and ear cartilage graft (OR, 2.20; 95% CI, 1.51-3.21; P < .001), hypertension (OR, 1.82; 95% CI, 1.29-2.57; P < .001), hypothyroidism (OR, 1.93; 95% CI, 1.10-3.39; P = .02), and tobacco use (OR, 1.65; 95% CI, 1.11-2.44; P = .01) (Table 3). Factors found to have a statistically significant association with readmission or return to the ED within 48 hours of surgery were postoperative bleeding (OR, 13.05; 95% CI, 4.24-40.16; P < .001), a diagnosis of neurologic disorder (OR, 4.11; 95% CI, 1.12-15.09; P = .03), and alcohol use (OR, 7.70; 95% CI, 1.55-38.21; P = .01) (Table 4).

Table 3. Risks Factors for Postoperative Admission^a.

Risk Factor	Total, No. (n = 2175)	Hospital Admissions, No. (%)	Univariable Analysis		Multivariable Analysis
			OR (95% CI)	P Value	OR (95% CI)	P Value
Age, y
<40	74	≤10 (≤13.5)	1 [Reference]		1 [Reference]
41-60	395	57 (14.4)	2.95 (1.04-8.40)	.04	2.80 (0.95-8.26)	.06
61-75	674	71 (10.5)	2.06 (0.73-5.81)	.17	1.65 (0.57-4.82)	.36
>75	772	67 (8.7)	1.66 (0.59-4.70)	.34	1.34 (0.46-3.91)	.60
Race
White	1705	185 (10.9)	1 [Reference]		1 [Reference]
Black	≤10	≤10 (≤0.5)	8.22 (1.15-58.68)		10.30 (0.83-128.01)	.07
Hispanic	71	≤10 (≤14.1)	0.90 (0.41-1.99)	.79	0.68 (0.30-1.59)	.38
Other	59	≤10 (≤16.9)	0.55 (0.17-1.78)	.32	0.65 (0.19-2.20)	.49
Adjacent tissue transfer
No	1514	138 (9.1)	1 [Reference]		1 [Reference]
Yes	401	61 (15.2)	1.79 (1.29-2.47)	<.001	1.88 (1.33-2.67)	<.001
Ear cartilage graft
No	1650	149 (9.0)	1 [Reference]		1 [Reference]
Yes	265	50 (18.9)	2.34 (1.65-3.33)	<.001	2.20 (1.51-3.21)	<.001
Peripheral vascular disease
No	1865	187 (10.0)	1 [Reference]		1 [Reference]
Yes	50	12 (24.0)	2.83 (1.46-5.52)	.002	1.56 (0.74-3.28)	.24
Hypertension
No	1210	96 (7.9)	1 [Reference]		1 [Reference]
Yes	705	103 (14.6)	1.99 (1.48-2.67)	<.001	1.82 (1.29-2.57)	<.001
Hypothyroidism
No	1804	177 (9.8)	1 [Reference]		1 [Reference]
Yes	111	22 (19.8)	2.27 (1.39-3.72)	.001	1.93 (1.10-3.39)	.02
Obesity
No	1842	186 (10.1)	1 [Reference]		1 [Reference]
Yes	73	13 (17.8)	1.93 (1.04-3.58)	.04	1.72 (0.83-3.56)	.14
Tobacco use
No	1599	141 (8.8)	1 [Reference]		1 [Reference]
Yes	316	58 (18.4)	2.32 (1.67-3.24)	<.001	1.65 (1.11-2.44)	.01

Abbreviation: OR, odds ratio.

^aFactors not listed were not statistically significant on univariable analysis.

Table 4. Risk Factors for Readmission or ED Visit Within 48 Hours of Surgery^a.

Risk Factor	Total	Readmissions or ED Visits, No. (%)	Univariable Analysis		Multivariable Analysis
			OR (95% CI)	P Value	OR (95% CI)	P Value
Postoperative bleeding
No	2145	33 (1.5)	16.00 (6.14-41.72)	<.001	13.05 (4.24-40.16)	<.001
Yes	30	≤10 (≤33.3)
Ear cartilage graft
No	1889	29 (1.5)	2.32 (1.12-4.82)	.02	2.04 (0.89-4.66)	.09
Yes	286	≤10 (≤3.5)
Neurologic disorder
No	2126	36 (1.7)	3.79 (1.13-12.74)	.03	4.11 (1.12-15.09)	.03
Yes	49	≤10 (≤20.4)
Renal failure
No	2101	35 (1.7)	3.37 (1.17-9.75)	.03	2.73 (0.84-8.84)	.10
Yes	74	≤10 (≤13.5)
Alcohol use
No	2153	37 (1.7)	5.72 (1.29-25.36)	.02	7.70 (1.55-38.21)	.01
Yes	22	≤10 (≤45.5)

Abbreviation: ED, emergency department; OR, odds ratio.

^aFactors not listed were not statistically significant on univariable analysis. “No” is the Reference category for all risk factors.

Discussion

Complication rates of 2175 patients undergoing paramedian forehead flap reconstruction for facial cancers were analyzed in this database study. To our knowledge, this is the largest cohort of forehead flap patients in the literature. Infection was the most common complication (2.9%), followed by bleeding (1.4%). Postoperative DVT rates were low (≤0.5%). In comparison, Little et al¹⁰ performed a retrospective review of 205 patients undergoing forehead flap surgery for nasal reconstruction and found a 31.7% major complication rate and a 50.8% minor complication rate. Major complications included flap necrosis, nasal obstruction, and alar notching. In that study, the authors had access to individual patient medical records and were able to track patients through postoperative clinic visits for 4 months. Detailed documentation allowed them to capture the nuances of each complication type and severity. In contrast, the present database study only captured complications coded through return visits to the ED, hospital, or ambulatory surgery center, which may explain the lower complication rates.

With regard to DVT rates, we found that this was a rare complication even in this cohort of patients with confirmed cutaneous malignancies undergoing a major surgery. Malignancy is regarded as an independent risk factor for developing VTE, including DVT. While some cancers, such as pancreatic, brain, and bone, are associated with higher risks of VTE than others, not all risk stratification models for VTE prophylaxis account for this difference.^19,20,21 For example, the Caprini score^21,22 is a validated model commonly used to stratify patients based on their risk factors for developing VTE. In this scoring system from 1 to 10, patients are given 2 points if they have an active or past diagnosis of cancer, and the resulting score is used to determine an appropriate prophylaxis regimen. This score can be very important because the prophylaxis regimen it dictates ranges widely from simple ambulation to therapeutic heparin.^21,22 Still, some have suggested that risk stratification strictly by Caprini scores should be handled differently in patients undergoing head and neck surgery.^23,24

Rudy et al²¹ recently performed a retrospective population cohort study analyzing the risk of VTE specifically in patients with keratinocyte carcinoma, including basal cell and squamous cell carcinoma. They concluded that there is no increased risk for VTE in patients with keratinocyte carcinoma compared with a control cohort. Their study design and conclusions are highly relevant to the present patient population, all of whom had a diagnosis of cutaneous cancer, most of which was caused by keratinocyte carcinoma. Indeed, the extremely low rate of DVT in the present study (≤0.5%) further reinforces the conclusion that patients with cutaneous cancer are likely not in the same risk category for developing VTE as patients with solid tumors. In turn, for patients who receive forehead flap surgery for cutaneous cancer, the surgeon should recognize that DVT is a much more rare complication than postoperative bleeding. As such, discretion should be used when deciding to administer anticoagulants to low- to medium-risk patients prior to surgery.

Patients in the present study were admitted to the hospital after surgery at a rate of 10.3%, while the rate of readmission within 30 days of surgery was 2.5%. Although we found no forehead flap statistics in the literature for direct comparison, Ryan et al²⁵ found a 3.2% hospital readmission rate for patients undergoing head and neck free flap procedures, and Mioton et al²⁶ report a 1.9% hospital readmission rate following outpatient plastic surgery. After performing multivariate analysis, we found that undergoing adjacent tissue transfer and ear cartilage graft concurrently with forehead flap surgery was significantly associated with increased risk for hospital admission. Adjacent tissue transfers, such as V-to-Y advancement flap, W-plasty, or even mucosal rotational flap, are routinely performed in conjunction with a forehead flap to help cover large cutaneous deficits and are instrumental in reconstructing full-thickness nasal deficits.^9,27 Ear cartilage grafts are commonly harvested to help add structural support to the forehead flap during nasal reconstruction.²⁸ Both of these procedures not only introduce an additional surgical site but may also extend the operating time. Moreover, these surgical sites often result in additional wound care tasks, which can prove burdensome for the patient, thereby prompting hospital admission for a period of observation.

Even though ear cartilage grafting was associated with significantly increased rates of hospital admission, it was not associated with rates of immediate readmissions or ED visits within 48 hours of surgery, based on multivariate analysis. This seems logical because complications related to ear cartilage grafting, such as malposition, extrusion, or graft necrosis, usually manifest on the order of days rather than hours postoperatively.²⁹ Likewise, tobacco use has been widely shown to have a negative effect on wound healing, which can lead to partial flap necrosis as a delayed complication.¹⁰ Indeed, our multivariate analysis found tobacco use was significantly associated with rate of admission after surgery, but not with rate of immediate readmission or ED visit within 48 hours postoperatively. In contrast, postoperative bleeding is a complication that would be expected to manifest quickly and prompt the patient to seek immediate medical attention, especially given that the forehead flap pedicle has a tendency to bleed. Indeed, bleeding was found to have 13 times greater odds of immediate readmission or ED visit within 48 hours of surgery than nonbleeding.

This database study provides statistical outcomes for a large cohort of patients undergoing paramedian forehead flap surgery. The large number of patients included in the data set is drawn from multiple states spanning different practice locations, institutions, and surgeons. Taken together, the statistical makeup of this study cohort is reflective of all patients receiving forehead flap surgery, and arguably yields more accurate results than single-institution and single-surgeon retrospective studies. Additionally, this study analyzes the association of several factors with surgical outcomes not previously studied, such as concurrent procedures and rate of DVT. The ability to study the rate of ED visits and hospital readmissions helps frame the management of those surgical complications, and how they affect patients in the immediate postoperative period. In an age where hospital readmissions are used as an important marker to judge the quality of patient care and determine reimbursement, data regarding surgery-specific readmission rates are useful for preoperative counseling and planning of postoperative care. The low rates of complications and readmissions found in this study suggest that forehead flaps can be routinely performed in the outpatient setting without the need for planned admission.

Limitations

The database from which this study is derived contributes multiple strengths but is also a source of limitations. Because the data are dependent on the ICD-9 and CPT codes recorded by health care professionals, the accuracy of the outcomes is based on the initial documentation of these codes. It is possible there were errors in coding and omissions of certain diagnoses for which no appropriate codes existed. Even though all codes are tallied equally in a database, the complications they reflect exist on a spectrum of severities that are not captured. The lack of descriptive documentation means that the nuances of these complications cannot be integrated into the data analysis. For example, postoperative infection is a complication that can be managed with outpatient antibiotics alone, or it may require readmission and return to the operating room for debridement. The fact that these different outcomes cannot be definitively separated based on the available coding confounds the data. Finally, because this study searched for the CPT code for forehead flap in the ambulatory surgery database, these data do not include patients who underwent forehead flap surgery in the inpatient setting. This potentially leads to the loss of more seriously ill patients who were preadmitted before scheduled surgery. Without these acute cases in the cohort, the complication and readmission rates may be falsely low. We recommend that these results be interpreted in reference to patients who have scheduled surgery in the outpatient setting.

Conclusions

The most common complication of paramedian forehead flap reconstruction is infection, followed by bleeding; DVT is a rare complication in patients undergoing paramedian forehead flap reconstruction. Because bleeding is a more common complication in this patient population, discretion should be used when deciding to administer anticoagulant drugs to low- to medium-risk patients prior to surgery.

Supplement.

eTable 1. Relevant Codes Used in Database