Complications of the Pectoralis Major Myocutaneous Flap in Head and Neck Reconstruction Jatin P. Shah, MD, FACS,Vijay Haribhakti, MD, Thom R. Loree, MD, Perry Sutaria, as, NewYork, NewYork A retrospective review of the complications in 211 patients undergoing pectoralis major myocutaneous flap reconstruction is presented. The flap was used for mucosal lining of the oral cavity or oropharynx in 109 patients, for pharyngoesophageal reconstruction in 44, for skin coverage in 47, and for other locations in 14 patients. Flap-related complications developed in 63% of the patients. These included flap necrosis, suture line dehiscence, fistula formation, infection, and hematoma. Analysis of risk factors for the development of flap complications showed the following factors to be significant: age over 70; female gender; homographic overweight; albumin less than 4 g/dL; use of the flap in reconstruction of the oral cavity after major glossectomy; and presence of other systemic diseases. The median length of hospitalization for those developing complications was 33 days compared with 16 days for those who did not develop any complications. Thirty-five (26%) of the 135 patients developing complications required reoperation and only 2 among these required a second flap. Similarly, only 13 of the 61 patients who developed fistulas required surgical closure.
ince its initial description by Ariyan [1,2] in 1979, the pectoralis major myocutaneous flap has proven to be a "workhorse" in reconstructive surgery of the head and neck [3,4]. The flap is versatile and has been used in many ingenious ways over the past 11 years. The applicability of this flap in reconstructive surgery of the head and neck is now universally accepted, and the flap has proven to be a definite advance in reconstructive surgery over previously employed methods. Sufficient experience has now been obtained to study the applicability, versatility, limitations, and complications of this flap. This article, however, presents a review of only the complications in a consecutive series of 211 patients undergoing pectoralis
S
From the Head and Neck Service, Memorial Sloan-Ketteriag Cancer Center, New York, New York. Requests for reprints should be addressed to Jatin P. Shah, MD, 1275 York Avenue, New York, New York 10021. Presented at the 36th Annual Meeting of the Society of Head and Neck Surgeons, Washington, DC, May 19-22, 1990.
352
THE AMERICAN JOURNAL OF SURGERY
major myocutaneous flap reconstruction on one service over the last 10 years. The data are analyzed to identify various risk factors leading to the development of significant complications. PATIENTS AND METHODS In a 10-year period from 1979 to 1989, 214 pectoralis major myocutaneous flap reconstructions were performed on 211 patients on the Head and Neck Service at Memorial Sloan-Kettering Cancer Center in New York. A retrospective review of the case records was undertaken to retrieve the clinical information, details of the operative procedure, and immediate and delayed postoperative complications. Nomographic characters of the patients and their physiologic and biochemical profiles, as well as associated conditions, were all carefully reviewed. The collected data were subjected to computer analysis for statistical significance using Fisher's exact test. RESULTS There were 144 men and 67 women. Their ages ranged from the second to the ninth decades, with the majority in the sixth and seventh decades. The sites of repair following excision of the primary tumor are shown in Figure 1. The oral cavity and the pharynx were the most common primary sites. Eighty-seven percent of the patients had squamous carcinoma of the upper aerodigestive tract. More than half (52%) of the patients underwent surgery for recurrent disease. Ninety-five of the 102 (93%) previously untreated patients had stage III or IV disease (American Joint Committee on Cancer staging-1988). One hundred nine (51%) flaps were used for mucosal lining of the oral cavity or oropharynx. Forty-four (20%) were used for pharyngoesophageal reconstruction, of which 35 were used as a patch and 9 were shaped into tubes for circumferential pharyngeal repair. Forty-seven (22%) flaps were used for skin coverage, and 7 (3%) had two skin islands on the same paddle for through-andthrough defects. In two patients, only muscle and subcutaneous fat were elevated to provide bulk. In three patients, only the muscle was used to provide protection for an exposed carotid artery, and in two additional patients, the muscle was used to cover exposed bone after temporal bone resection. Postoperative complications: Under this category, all complications occurring between the time of surgery and discharge from the hospital were listed. Regardless of the relative importance of the complications in each case, the numbers reported in this study include all complications encountered. The complications were grouped into two broad categories: those directly related to the pector-
VOLUME 160
OCTOBER 1990
COMPLICATIONS OF MYOCUTANEOUS FLAP RECONSTRUCTION
alis major myocutaneous flap (flap complications) and other complications. Only 47 patients (22%) had a completely uneventful recovery. Two patients (1%) died, one each on postoperative days 2 and 6, due to cardiac causes. One hundred thirty-five patients (63%) developed flap complications and 30 (14%) had other complic~/tionsnot related to the flap. Flap-related complications are listed in Table I. Several patients had multiple flap-related complications, e.g., flap necrosis, dehiscence, infection, and fistula formation. Several risk factors were analyzed in each case for their correlation with the flap complications. The risk factors studied were as follows: (1) Age, sex, height, and weight, as well as nomograms to define "overweight," "desirable weight," and "underweight"[5]; (2) Serum albumin values categorized in three groups: (a) less than 4 g/alL, (b) 4 to 5 g/dL, and (c) more than 5 g/dL; (3) Other diseases grouped as a broad category. Systemic diseases such as diabetes mellitus, hypertension, atherosclerotic heart disease, peripheral vascular disease, renal failure, and collagen diseases were included in this category, Since a number of conditions coexisted, it was difficult to judge the individual significance of each; (4) Site of primary tumor; (5) T stage of the primary lesion; (6) Extent of surgery; (7) Disposition of the flap (whether employed as mucosal lining or skin cover); and (8) The influence of previous treatment (surgery/radiation/chemotherapy). Flap necrosis. Sixty-nine patients (32%) developed varying degrees of flap necrosis. Of these, 7 (3%) had total necrosis. None of these patients had any intraoperative accidents, injury, difficulties, or demonstrable vascular anomaly that could have accounted for total necrosis. One male patient was a heavy smoker with a history of severe peripheral vascular disease. He developed dry gangrene of the entire flap. Histo!ogic study of the vascular pedicle showed total occlusion due to small vessel athero, sclerosis. One female patient with systemic lupus erythematosus developed delayed total necrosis on the fourth postoperative day. Another male patient who required the flap to cover an extensive defect after temporal bone resection developed progressive necrosis of the entire skin paddle over a period of 10 days. No contributing factor could be identified in this patient. The remaining four patients developed necrosis secondary to constriction of the vascular pedicle produced by the umbilical tape used to secure the tracheostomy tube. Management of total flap necrosis in these seven patients required surgical debridement or flap excision and secondary repair of the defect by other flaps and skin grafts in five and conservative debridement without surgical intervention in two. In 62 patients (29%), partial necrosis of the flap developed. Two patterns of flap loss were observed. Distal flap loss accounted for 22% (48 patients) and marginal flap loss occurred in the remaining 7% (14 patients). Distal flap loss was of partial thickness in 39 patients and full thickness in 9 patients. The high-risk factors of significance for flap necrosis are listed in Table II. Development of flap necrosis had a
Oral c a v i t y 86
)ther 9 4% Parotid 10 5%
Skin 15 7%
Oropharyl 27%
qypopharyr/x 36 17%
Figure 1. Sites of repair by pectoralis major myocutaneous flaps.
TABLE I
Flap-Related Complications Patients Flap necrosis Total Partial Fistula Dehiscence Infection Hematoma
n
%
69 7 62 61 56 51 14
32 3 29 29 26 24 7
significant impact upon the length of hospitalization (Table II). Fistula formation. Sixty-one patients (29%) developed fistulas. However, 48 of these closed spontaneously with conservative management. In 13 patients, a second surgical procedure was necessary to close the fistula. Development of fistulas significantly prolonged hospitalization (Table III). The significant risk factors for fistula formation included major resection in the oral cavity or laryngopharynx, a T4 primary tumor, use of the flap for mucosal lining, and the presence of other systemic diseases. Dehiscence. Fifty-six patients (26%) developed dehiscence of the flap suture line and 44 of them went on to develop other flap-related complications. The remaining 12 patients with only dehiscence of the suture line also had significant prolongation of their hospital stay (Table III). Significant risk factors for wound dehiscence includ, ed female gender, major resections for oral tumors, mandible resection, the presence of other systemic diseases, and use of the flap for mucosal lining. Infection. Fifty-one patients (24%) developed wound infection. All but nine were associated with other flap complications. Wound infection clearly prolonged hospital stay (Table III). Significant risk factors for this complication included overweight, female gender, the presence of other systemic diseases, T4 primary lesions, and use of the flap after major mandible resection. Hematoma. Wound hematoma occurred in 14 patients, 6 of whom required reoperation and evacuation.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 160
OCTOBER 1990
353
SHAH ET AL
TABLE II Significant High-Risk Factors for Flap Necrosis
(p
ince its initial description by Ariyan [1,2] in 1979, the pectoralis major myocutaneous flap has proven to be a "workhorse" in reconstructive surgery of the head and neck [3,4]. The flap is versatile and has been used in many ingenious ways over the past 11 years. The applicability of this flap in reconstructive surgery of the head and neck is now universally accepted, and the flap has proven to be a definite advance in reconstructive surgery over previously employed methods. Sufficient experience has now been obtained to study the applicability, versatility, limitations, and complications of this flap. This article, however, presents a review of only the complications in a consecutive series of 211 patients undergoing pectoralis
S
From the Head and Neck Service, Memorial Sloan-Ketteriag Cancer Center, New York, New York. Requests for reprints should be addressed to Jatin P. Shah, MD, 1275 York Avenue, New York, New York 10021. Presented at the 36th Annual Meeting of the Society of Head and Neck Surgeons, Washington, DC, May 19-22, 1990.
352
THE AMERICAN JOURNAL OF SURGERY
major myocutaneous flap reconstruction on one service over the last 10 years. The data are analyzed to identify various risk factors leading to the development of significant complications. PATIENTS AND METHODS In a 10-year period from 1979 to 1989, 214 pectoralis major myocutaneous flap reconstructions were performed on 211 patients on the Head and Neck Service at Memorial Sloan-Kettering Cancer Center in New York. A retrospective review of the case records was undertaken to retrieve the clinical information, details of the operative procedure, and immediate and delayed postoperative complications. Nomographic characters of the patients and their physiologic and biochemical profiles, as well as associated conditions, were all carefully reviewed. The collected data were subjected to computer analysis for statistical significance using Fisher's exact test. RESULTS There were 144 men and 67 women. Their ages ranged from the second to the ninth decades, with the majority in the sixth and seventh decades. The sites of repair following excision of the primary tumor are shown in Figure 1. The oral cavity and the pharynx were the most common primary sites. Eighty-seven percent of the patients had squamous carcinoma of the upper aerodigestive tract. More than half (52%) of the patients underwent surgery for recurrent disease. Ninety-five of the 102 (93%) previously untreated patients had stage III or IV disease (American Joint Committee on Cancer staging-1988). One hundred nine (51%) flaps were used for mucosal lining of the oral cavity or oropharynx. Forty-four (20%) were used for pharyngoesophageal reconstruction, of which 35 were used as a patch and 9 were shaped into tubes for circumferential pharyngeal repair. Forty-seven (22%) flaps were used for skin coverage, and 7 (3%) had two skin islands on the same paddle for through-andthrough defects. In two patients, only muscle and subcutaneous fat were elevated to provide bulk. In three patients, only the muscle was used to provide protection for an exposed carotid artery, and in two additional patients, the muscle was used to cover exposed bone after temporal bone resection. Postoperative complications: Under this category, all complications occurring between the time of surgery and discharge from the hospital were listed. Regardless of the relative importance of the complications in each case, the numbers reported in this study include all complications encountered. The complications were grouped into two broad categories: those directly related to the pector-
VOLUME 160
OCTOBER 1990
COMPLICATIONS OF MYOCUTANEOUS FLAP RECONSTRUCTION
alis major myocutaneous flap (flap complications) and other complications. Only 47 patients (22%) had a completely uneventful recovery. Two patients (1%) died, one each on postoperative days 2 and 6, due to cardiac causes. One hundred thirty-five patients (63%) developed flap complications and 30 (14%) had other complic~/tionsnot related to the flap. Flap-related complications are listed in Table I. Several patients had multiple flap-related complications, e.g., flap necrosis, dehiscence, infection, and fistula formation. Several risk factors were analyzed in each case for their correlation with the flap complications. The risk factors studied were as follows: (1) Age, sex, height, and weight, as well as nomograms to define "overweight," "desirable weight," and "underweight"[5]; (2) Serum albumin values categorized in three groups: (a) less than 4 g/alL, (b) 4 to 5 g/dL, and (c) more than 5 g/dL; (3) Other diseases grouped as a broad category. Systemic diseases such as diabetes mellitus, hypertension, atherosclerotic heart disease, peripheral vascular disease, renal failure, and collagen diseases were included in this category, Since a number of conditions coexisted, it was difficult to judge the individual significance of each; (4) Site of primary tumor; (5) T stage of the primary lesion; (6) Extent of surgery; (7) Disposition of the flap (whether employed as mucosal lining or skin cover); and (8) The influence of previous treatment (surgery/radiation/chemotherapy). Flap necrosis. Sixty-nine patients (32%) developed varying degrees of flap necrosis. Of these, 7 (3%) had total necrosis. None of these patients had any intraoperative accidents, injury, difficulties, or demonstrable vascular anomaly that could have accounted for total necrosis. One male patient was a heavy smoker with a history of severe peripheral vascular disease. He developed dry gangrene of the entire flap. Histo!ogic study of the vascular pedicle showed total occlusion due to small vessel athero, sclerosis. One female patient with systemic lupus erythematosus developed delayed total necrosis on the fourth postoperative day. Another male patient who required the flap to cover an extensive defect after temporal bone resection developed progressive necrosis of the entire skin paddle over a period of 10 days. No contributing factor could be identified in this patient. The remaining four patients developed necrosis secondary to constriction of the vascular pedicle produced by the umbilical tape used to secure the tracheostomy tube. Management of total flap necrosis in these seven patients required surgical debridement or flap excision and secondary repair of the defect by other flaps and skin grafts in five and conservative debridement without surgical intervention in two. In 62 patients (29%), partial necrosis of the flap developed. Two patterns of flap loss were observed. Distal flap loss accounted for 22% (48 patients) and marginal flap loss occurred in the remaining 7% (14 patients). Distal flap loss was of partial thickness in 39 patients and full thickness in 9 patients. The high-risk factors of significance for flap necrosis are listed in Table II. Development of flap necrosis had a
Oral c a v i t y 86
)ther 9 4% Parotid 10 5%
Skin 15 7%
Oropharyl 27%
qypopharyr/x 36 17%
Figure 1. Sites of repair by pectoralis major myocutaneous flaps.
TABLE I
Flap-Related Complications Patients Flap necrosis Total Partial Fistula Dehiscence Infection Hematoma
n
%
69 7 62 61 56 51 14
32 3 29 29 26 24 7
significant impact upon the length of hospitalization (Table II). Fistula formation. Sixty-one patients (29%) developed fistulas. However, 48 of these closed spontaneously with conservative management. In 13 patients, a second surgical procedure was necessary to close the fistula. Development of fistulas significantly prolonged hospitalization (Table III). The significant risk factors for fistula formation included major resection in the oral cavity or laryngopharynx, a T4 primary tumor, use of the flap for mucosal lining, and the presence of other systemic diseases. Dehiscence. Fifty-six patients (26%) developed dehiscence of the flap suture line and 44 of them went on to develop other flap-related complications. The remaining 12 patients with only dehiscence of the suture line also had significant prolongation of their hospital stay (Table III). Significant risk factors for wound dehiscence includ, ed female gender, major resections for oral tumors, mandible resection, the presence of other systemic diseases, and use of the flap for mucosal lining. Infection. Fifty-one patients (24%) developed wound infection. All but nine were associated with other flap complications. Wound infection clearly prolonged hospital stay (Table III). Significant risk factors for this complication included overweight, female gender, the presence of other systemic diseases, T4 primary lesions, and use of the flap after major mandible resection. Hematoma. Wound hematoma occurred in 14 patients, 6 of whom required reoperation and evacuation.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 160
OCTOBER 1990
353
SHAH ET AL
TABLE II Significant High-Risk Factors for Flap Necrosis
(p
