- Email: [email protected]
Prevention of postoperative pericardial adhesions with hydrophilic polymer solutions
JOURNAL
OF SURGICAL
45, 44-49 (1988)
RESEARCH
Prevention
of Postoperative Pericardial Adhesions Hydrophilic Polymer Solutions
with
D. A. DUNCAN, M.D., Y. YAACOBI, M.D., PH.D.,* E. P. GOLDBERG, PH.D.,* M. MINES, M.S., D. O’BRIEN, R.N., F. CONGDON, B.S., AND M. J. CARMICHAEL, M.D. Department of Surgery, University of Florida, Gainesvile, Florida 32610; *Biomedical Engineering Center, Department of Materials Science & Engineering, MAE 217, University of Florida, Gainesville, Florida 3261 I; and University Hospital and Trauma Center, Jacksonville, Florida Presented at the Annual Meeting of the Association for Academic Surgery, Orlando, Florida, November 1-4, 1987 Pericardial adhesions subject patients requiring cardiac reoperation to potential injuries of the heart, great vessels,and extracardiac grafts during restemotomy. We evaluated polyvinylpyrrolidone (PVP) and a methylcellulose derivative (MCD) as intraoperative irrigating solutions in the prevention of postoperative pericardial adhesions. Fifteen dogs weighing 15 to 20 kg were divided into three equal cohorts and subjected to left thoracotomy with pericardiotomy. Prior to surgical manipulation, the pericardial cavity was irrigated with either PVP, MCD, or Ringer’s lactate (RL). Serosal injury simulating intraoperative trauma was induced by gauze sponge abrasion of the epicardium and inner surface of the pericardium and by allowing desiccation of serosal surfaces for 30 min. The pericardial cavity was evaluated for adhesion formation 6 weeks postoperatively by reoperation. Two independent observers, unaware of the study solution, evaluated the extent and severity of pericardial adhesions on a O-4 scale. All dogs in the RL control group had surgically significant adhesions and a mean adhesion score of 3.2 f 1.1. In contrast, no PVP- or MCD-treated dog had surgically significant adhesions. The mean adhesion scores were 0.2 + 0.4 for PVP and 0.5 f 0.7 for MCD. Our results indicate that PVP (P < 0.004) and MCD (P < 0.024) significantly reduce pericardial adhesion formation when compared to RL. Clinical application of PVP or MCD for the prevention of pericardial adhesions should reduce morbidity and mortality of cardiac reoperation. o 1988 Academic press, IX.
In view of the markedly increasing incidence of cardiac reoperations and potentially serious complications related to pericardial adhesions, intensive efforts to prevent adhesion formation are necessary. We have been studying the effects of hydrophilic polymer solutions on the formation of postoperative peritoneal adhesions and have demonstrated promising results with polyvinylpyrrolidone
INTRODUCTION
The incidence of cardiac reoperation continues to steadily increase with repeat coronary artery surgery comprising the majority of reoperations today. Postoperative pericardial adhesions subject patients undergoing repeat cardiac surgery to substantial risks. Potential injury to the heart, great vessels, and extracardiac grafts during restemotomy, as well as increased operative time, increases morbidity and mortality. Restemotomy is associated with a 2-6% incidence of major vascular injury. Dobell and Iain reported a 37% mortality in patients experiencing major hemorrhage during resternotomy; in addition, injury to aortocoronary artery grafts is reported to have a 50% mortality [2, 71. Significant morbidity, mortality, and tedious time-consuming dissection command respect of the surgeon for pericardial adhesions during cardiac reoperation. 0022-4804/88 $1.50 Copyright 0 1988 by Academic Press, Inc. All rigbls of reproduction in any form reserved.
(PVP) 151.
The following study was designed to evaluate the effects of PVP and a methylcellulose derivative (MCD) as irrigating solutions in the prevention of postoperative pericardial adhesions. MATERIALS
AND METHODS
Fifteen mongrel dogs weighing between 15 and 20 kg were maintained and cared for in a facility fully accredited by the American Association for Laboratory Animal Care. Man44
DUNCAN
ET AL.: PREVENTION
agement and surgery of all animals were in accordance with the Federal Animal Welfare Act and the NZH Guide for Care and Use of Laboratory Animals. Thirty percent PVP and 1 percent MCD were used as irrigating solutions for the experimental groups. Animals in the control group were irrigated with Ringer’s lactate (RL) solution. Fifteen dogs were divided into three equal groups according to the solution used for pericardial irrigation. A single intravenous dose of Biotal(O.04 mg/kg) was given to induce anesthesia. General anesthesia was maintained with 2.0% Ethrane. The animals were intubated by direct laryngoscopy and mechanical ventilation was maintained with a volume-limited ventilator at 8- 10 breaths/ min, 15 cc/kg tidal volume, and 0.2 1 FIO2. After each animal was prepped and draped in a sterile fashion, a 12-cm left thoracotomy incision was performed at the fifth intercostal space. Prior to manipulation of internal organs, all instruments, gauze sponges, and gloves were bathed in the study solution. A 6-cm pericardiotomy was placed 3 cm anterior to the left phrenic nerve. This was immediately followed by irrigation of the pericardial sac with 20 cc of the study solution. The epicardium and the inner surface of the pericardium were abraded with 20 strokes of a gauze sponge to induce a punctated hemorrhagic lesion. The inner surface of the pericardium was irrigated three times with 20 cc of solution: immediately following pericardiotomy, 15 min following pericardiotomy, and prior to pericardial closure. After the epicardium and inner surface of the pericardium had been exposed to air for 30 min, to simulate intraoperative desiccation, the pericardium was closed loosely with 2-O silk sutures. Thoracotomy closure was performed in the usual fashion including aspiration of air and fluid from the left chest. Intercostal nerve blocks were induced with 1% xylocaine and l/ 100,000 epinephrine. Flocillin and Demerol were given for 7 postoperative days, and all dogs received regular diets. Reoperation for evaluation of pericardial adhesions was performed 6 weeks following
OF PERICARDIAL
ADHESIONS
45
initial surgery. Under general anesthesia all animals were subjected to bilateral thoracotomies. The pericardial cavity was evaluated for adhesions by two observers unaware of the study solution. A scale of O-4 was used to score adhesions of the abraded surface, unabraded surface, and suture line: O-no adhesions 1-filmy adhesions with easily identifiable plane 2-mild adhesions with freely dissectable plane 3-moderate adhesions with difficult dissection of plane 4-dense adhesions with nondissectable plane Photographs were taken of the pericardial cavity and adhesions (if present) to allow additional comparison and documentation (Figs. l-5). All animals were sacrificed and biopsies of abraded and unabraded epicardium and abraded and unabraded pericardium underwent histologic evaluation. The incidence and severity of adhesions in animals treated with hydrophilic polymer solutions (PVP and MCD) were compared with those of the control group treated with RL. Statistical analysis was performed with Fisher’s exact test. RESULTS
Pericardial adhesion scores are summarized in Table 1. There was no difference in
FIG. 1. PVP Grade 0. No adhesions were present at autopsy in this animal with polyvinylpyrrolidone solution.
46
JOURNAL
OF SURGICAL
RESEARCH:
VOL. 45, NO. 1, JULY 1988
FIG. 2. MCD Grade 0. No adhesions were present at autopsy in this animal irrigated with a methylcellulose derivative solution.
FIG. 4. RL Grade 4. Dense adhesions without dissectable planes were present at autopsy in this animal irrigated with Ringers lactate.
adhesion formation between the abraded surface and suture line. The RL-irrigated control group had adhesion scores >3 in all but one dog. None of these dogs were free of pericardial adhesions and the mean adhesion score was 3.2 + 1.1. Four of five PVP-treated dogs, on the contrary, showed no postoperative adhesion. The mean adhesion score was 0.2 + 0.4. An iatrogenic foreign body (sponge) was found in the lone PVP-treated animal who developed a mild pericardial adhesion. The MCD-treated group showed marked reduction of pericardial adhesion formation and had a mean adhesion score of 0.5 + 0.7. No animal in the MCD group developed a surgically significant adhesion (>2 adhesion score). A marked reduction in pericardial adhesion formation was shown when
comparing PVP and RL (P < 0.004) and when comparing MCD and RL (I’ < 0.024). No statistical difference existed between PVP and MCD (P < 0.5); however, there was a lower mean adhesion score with PVP (0.2) than with MCD (0.5). Histologic sections of epicardium and pericardium sent for evaluation of thickness, chronic inflammation, fibrosis, and neovasculasity revealed no trends or statistical differences in any parameter between PVP, MCD, and RL cohorts.
FIG. 3. RL Grade 2. Mild adhesions with freely dissectable planes were present at autopsy in this animal irrigated with Ringers lactate.
FIG. 5. RL Grade 4. Dense adhesions without dissectable planes were present at autopsy in this animal irrigated with Ringers lactate.
DISCUSSION
Incidence and Management of Pericardial Adhesions Cardiac reoperation constitutes 10% of heart surgery presently performed and its in-
DUNCAN
ET AL.: PREVENTION
OF PERICARDIAL
47
ADHESIONS
TABLE 1 POSTOPERATIVE PERICARDIAL ADHESIONS U
A
SL
C
RL,= RL2 RLS RL., RLS
0 0 0 0 0
4 2 3 2 4
4 I 4 4 4
4 1.5 3.5 3 4
Mean
0
3
3.4
3.2
U
A
SL
C
MCD, MCD2 MCDs MCD., MCDS
0 0 0 0 0
1 2 0 0 0
1 0 0 1 0
1 1 0 0.5 0
PVP, PVPZ PVPS PVP, PVPSh
0 0 0 0 0
0 0 0 0 1
0 0 0 0 1
0 0 0 0 1
Mean
0
0.6
0.4
0.5
Mean
0
0.2
0.2
0.2
U
Note. RL, Ringers lactate; MCD, methylcellulose derivative, PVP, polyvinylpyrrolidone; abraded; SL, suture line; C, combined abraded and suture line. a Sacrificed at 12 weeks because dog had puppies. b Two sponges inadvertently left in hemithorax.
A
SL
C
U, unabraded; A,
cidence continues to increase [ 151. The pri- instead of prevention. Dobel and Iain [2] mary indication for reoperation is coronary have outlined several steps for safer restemorevascularization [7]. Other frequent indicatomy: appreciate that a high risk situation tions include prosthetic valve failure and de- exists, review lateral chest radiographs, exfinitive cardiac repair following staged or pose the femoral artery and vein for potential palliative cardiac procedures in children with circulatory control with cardiopulmonary congenital heart disease. Following cardiac bypass, cautious reentry, retract sternal surgery, adhesions form between the epicar- tables upward, avoid premature placement dium and surrounding structures including of the sternal retractor, deflate lungs, and use the pericardium, mediastinal tissues, pleura, a low power oscillating saw [2, 71. Although and sternum. Reoperation is prolonged by surgical techniques for reoperation have imdifficult sternal reentry, adhesion lysis, and proved, resternotomy remains a challenge increased hemorrhage. Restemotomy is as- for the cardiac surgeon. sociated with a 2-6s incidence of major vascular injuries and a 37% mortality when major hemorrhage occurs [2, 71. Injuries to Adhesion Formation the innominate vein, superior vena cava, A variety of mechanisms have been shown right atrium, right ventricle, pulmonary arto be important in adhesion formation. All tery, and aorta are not uncommon. Aortogenerally involve the healing response to secoronary grafts involving the right coronary rosal injury. Pericardial injury occurs with artery and internal mammary arteries are desiccation, abrasion, and foreign body reacparticularly vulnerable to injury [7]. Furthertion. The etiology of surgical adhesion formore, adhesions and fibrosis obscure anatomic structures making identification of mation has been extensively reviewed [ 1, 4, 5,9, 14, 151. coronary vessels difficult. Five sequential steps seem to be involved: High risk situations for restemotomy include multiple reoperations, previous as- (1) initial injury to the surface endothelium, (2) pouring out of exudate at the site of incending aortic aneurysm repairs, multiple valve disease, right atrial or ventricular en- jury, (3) coagulation of the exudate and filargement, right ventricular-pulmonary ar- brin formation, (4) fibrin adhesiveness caustery conduits, and previous mediastinitis [7]. ing adjacent structures to stick together, and The present approach to pericardial adhe- (5) organization with fibrous tissue formasions involves attempts at safe restemotomy tion.
48
JOURNAL
OF
SURGICAL
RESEARCH:
VOL.
45, NO.
I, JULY
1988
Prevention of Adhesion Formation
Hydrophilic Polymer Solutions
Although the prevention of pericardial adhesions has only recently been investigated, much has been learned in the prevention of postoperative adhesions from research in abdominal and opthalmic surgery. Present efforts to prevent postoperative pericardial adhesions center on three principles: (1) preventing unnecessary serosal injury; (2) primary pericardial closure, pericardial augmentation, or pericardial substitutes; and (3) local or systemic pharmacology [ 1 I]. These efforts to date have reported minimal success. This study did not attempt to address the question of pericardial closure or substitutes. Although our study was confined to the pericardial cavity, an understanding of the mechanism of hydrophilic polymer solutions and their use in previous peritoneal studies suggest that the results may be extrapolated to extrapericardial structures. The relationship of serosal injury to postoperative adhesion formation has been documented. Crush injuries, abrasions, foreign body reactions, desiccation, infection, dry gloves, and sponges have all been implicated in adhesion formation. Primary pericardial closure had been advocated; however, it has largely been abandoned due to kinking of grafts and disturbances of cardiac filling [7]. Pericardial meshing has been reported to show some success [ 111. Although some pericardial substitutes appear to reduce adhesion formation, their use has been associated with severe epicardial fibrosis creating difficulty identifying coronary vessels [3, lo]. Systemic agents such as promethazine, dexamethazone, urokinase, streptokinase, heparin, and dextran have had Limited success; however, each is associated with significant side effects [ 15, 161. Local application of Dextran 70, in the rabbit pericardial cavity, to separate serosal surfaces, has reportedly shown some beneficial effect [ 141. Hydrophilic polymer solution tissue coatings have been proven to be effective in animal studies for the prevention of peritoneal adhesions, but have not been reported in pericardial adhesion studies to date [5, 91.
Studies on the reduction of surgical adhesions have focused on preventing the process of adhesion formation following significant tissue injury. An important aspect of this study was the use of protective nonadhesive tissue coatings with a hydrophilic polymer solution. Such tissue coatings may be expected to reduce desiccation and prevent sticking of tissues. The objective is to prevent tissue damage during surgical manipulation rather than reverse a surgical injury. Goldberg et al. showed a marked reduction of peritoneal adhesions with PVP solutions [5]. In addition, it was pointed out that significant serosal injuries are induced by manipulative abrasions due to sponges and surgeons’ gloves. Electron microscopy showed a major reduction in tissue damage following the application of PVP solutions [5]. SUMMARY
Our results indicate that hydrophilic polymer solutions such as PVP and MCD significantly reduce pericardial adhesion formation when compared to controls. Clinical application of hydrophilic polymer solutions for the prevention of pericardial adhesions should reduce morbidity and mortality of cardiac reoperation. Hydrophilic polymer solutions have been subjected to extensive toxicologic evaluations and have been found to have no adverse effects; however, additional research is needed to determine the most efficacious hydrophilic polymers, optimal molecular weights, concentrations, osmolarities, pH’s, and body clearances. Bioacceptance, long term effects, and methods of application for hydrophilic polymer solutions need further investigation. REFERENCES 1. ClifT, W. J., Grobety, J., and Ryan, G. B. Postoperative pericardial adhesions: The role of mild serosal injury and spilled blood. J. Thoruc. Curdiovasc. Surg. 65: 1973. 2. Dobell, R. C., and Iain, A. K. Catastrophic hemor-
DUNCAN
3.
4. 5.
6. 7. 8.
ET AL.:
PREVENTION
rhage during redo stemotomy. Ann. Thorac. Surg. 37: 1984. Gallo, I., Artinano, E., Duran, C. G. Late clinical results with the use of heterologous pericardium for closure of the pericardial cavity. J. Thorac. Cardiovast. Surg. 89: 109, 1985. Gervin, A. S., Jacobs, G., Hofnagel, H., and Mason, K. Surgical trauma and pericardial fibrinolytic activity. Amer. Surg. 225, 1915. Goldberg, E. P., Sheets, J. W., and Habal, M. B. Peritoneal adhesions: Prevention with the use of hydrophilic polymer coatings. Arch. Surg. 115: 776, 1980. Londe, S., and Sugg, W. L. The challenge of reoperation in cardiac surgery. Ann. Thorac. Surg. 17: 157, 1974. Loop, F. D. Catastrophic hemorrhage during sternal reentry. Ann. Thorac. Surg. 37: 271, 1984. Math&n, S. R., Hong-De, W., Savvage, L. R., and Walker, M. W. Prevention of retrostemal adhesions after pericardiotomy. J. Thorac. Cardiovasc. Surg. 92: 92, 1986.
Mazuji, M., and Fadhh, H. A. Peritoneal adhesions: Prevention with povidone and dextran 75. Arch. Surg. 91: 872, 1965. 10. Meus, P. J., Wemly, J. A., Campbell, C. D., Takanaski, Y., Pick, R. L., Qui Zhao-Kun, and Replogle, R. L. Long-term evaluation of pericardial substitutes. J. Thorac. Cardiovasc. Surg. 85: 54, 1983. 11. Milgaltes, E., Gideon, V., Siberman, S., Appelbaum, Y., Shimon, D., Kopolevic, J., Cohen, D., Jones, H., Appelbaum, A., and Berman, J. Pericar9.
OF
PERICARDIAL
ADHESIONS
49
dial meshing: An effective method for prevention of pericardial adhesions and epicardial reaction after cardiac operations. J. Thorac. Cardiovasc. Surg. 90: 281, 1985. 12. Oyer, P. E., and Shumway, N. E. Again, via the median stemotomy. Arch. Surg. 109: 604, 1974. 13. Pucci, J. J., Walesby, R. K., Smith, E. J., Edmondson, S. J., Sapsford, R. N., Bentall, H. H., and Cleland, W. P. Reoperation for failed aorto-coronary bypass grafts. J. Cardiovasc. Surg. 23: 453, 1982. 14. Robison, R. J., Brown, J. W., Deschnes, W. P., Highes, B., and King, H. Prevention of pericardial adhesions with Dextran 70. Ann. Thorac. Surg. 37: 488, 1984.
15. Salm, T. J., Okike, 0. N., Massicano, T. N., Comp ton, C., and Espinoza, E. Prevention of postoperative pericardial adhesions. Arch. Surg. 121: 462, 1986.
Smith, L. 0. Prevention of surgically induced pericardial adhesions with combined dexamethasone and promethazine therapy. J. Flu. Med. Assoc. 55: 413, 1968. 17. Ungerlieder, R. M., Mills, N. L., and Wechsler, A. S. Left thoracotomy for reoperative coronary artery bypass procedures. Ann. Thorac. Surg. 40: 11, 1985. 18. Urschel, H. C., Razzok, M. A., and Gardner, M. Coronary artery bypass occlusion secondary to postcardiotomy syndrome. Ann. Thorac. Surg. 22: 528, 1976. 19. Youmans, C. R., White, J., and Derrick, J. R. The prevention of pleural and pericardial adhesions with silastic. J. Thorac. Cardiovasc. Surg. 55: 383, 1968. 16.
OF SURGICAL
45, 44-49 (1988)
RESEARCH
Prevention
of Postoperative Pericardial Adhesions Hydrophilic Polymer Solutions
with
D. A. DUNCAN, M.D., Y. YAACOBI, M.D., PH.D.,* E. P. GOLDBERG, PH.D.,* M. MINES, M.S., D. O’BRIEN, R.N., F. CONGDON, B.S., AND M. J. CARMICHAEL, M.D. Department of Surgery, University of Florida, Gainesvile, Florida 32610; *Biomedical Engineering Center, Department of Materials Science & Engineering, MAE 217, University of Florida, Gainesville, Florida 3261 I; and University Hospital and Trauma Center, Jacksonville, Florida Presented at the Annual Meeting of the Association for Academic Surgery, Orlando, Florida, November 1-4, 1987 Pericardial adhesions subject patients requiring cardiac reoperation to potential injuries of the heart, great vessels,and extracardiac grafts during restemotomy. We evaluated polyvinylpyrrolidone (PVP) and a methylcellulose derivative (MCD) as intraoperative irrigating solutions in the prevention of postoperative pericardial adhesions. Fifteen dogs weighing 15 to 20 kg were divided into three equal cohorts and subjected to left thoracotomy with pericardiotomy. Prior to surgical manipulation, the pericardial cavity was irrigated with either PVP, MCD, or Ringer’s lactate (RL). Serosal injury simulating intraoperative trauma was induced by gauze sponge abrasion of the epicardium and inner surface of the pericardium and by allowing desiccation of serosal surfaces for 30 min. The pericardial cavity was evaluated for adhesion formation 6 weeks postoperatively by reoperation. Two independent observers, unaware of the study solution, evaluated the extent and severity of pericardial adhesions on a O-4 scale. All dogs in the RL control group had surgically significant adhesions and a mean adhesion score of 3.2 f 1.1. In contrast, no PVP- or MCD-treated dog had surgically significant adhesions. The mean adhesion scores were 0.2 + 0.4 for PVP and 0.5 f 0.7 for MCD. Our results indicate that PVP (P < 0.004) and MCD (P < 0.024) significantly reduce pericardial adhesion formation when compared to RL. Clinical application of PVP or MCD for the prevention of pericardial adhesions should reduce morbidity and mortality of cardiac reoperation. o 1988 Academic press, IX.
In view of the markedly increasing incidence of cardiac reoperations and potentially serious complications related to pericardial adhesions, intensive efforts to prevent adhesion formation are necessary. We have been studying the effects of hydrophilic polymer solutions on the formation of postoperative peritoneal adhesions and have demonstrated promising results with polyvinylpyrrolidone
INTRODUCTION
The incidence of cardiac reoperation continues to steadily increase with repeat coronary artery surgery comprising the majority of reoperations today. Postoperative pericardial adhesions subject patients undergoing repeat cardiac surgery to substantial risks. Potential injury to the heart, great vessels, and extracardiac grafts during restemotomy, as well as increased operative time, increases morbidity and mortality. Restemotomy is associated with a 2-6% incidence of major vascular injury. Dobell and Iain reported a 37% mortality in patients experiencing major hemorrhage during resternotomy; in addition, injury to aortocoronary artery grafts is reported to have a 50% mortality [2, 71. Significant morbidity, mortality, and tedious time-consuming dissection command respect of the surgeon for pericardial adhesions during cardiac reoperation. 0022-4804/88 $1.50 Copyright 0 1988 by Academic Press, Inc. All rigbls of reproduction in any form reserved.
(PVP) 151.
The following study was designed to evaluate the effects of PVP and a methylcellulose derivative (MCD) as irrigating solutions in the prevention of postoperative pericardial adhesions. MATERIALS
AND METHODS
Fifteen mongrel dogs weighing between 15 and 20 kg were maintained and cared for in a facility fully accredited by the American Association for Laboratory Animal Care. Man44
DUNCAN
ET AL.: PREVENTION
agement and surgery of all animals were in accordance with the Federal Animal Welfare Act and the NZH Guide for Care and Use of Laboratory Animals. Thirty percent PVP and 1 percent MCD were used as irrigating solutions for the experimental groups. Animals in the control group were irrigated with Ringer’s lactate (RL) solution. Fifteen dogs were divided into three equal groups according to the solution used for pericardial irrigation. A single intravenous dose of Biotal(O.04 mg/kg) was given to induce anesthesia. General anesthesia was maintained with 2.0% Ethrane. The animals were intubated by direct laryngoscopy and mechanical ventilation was maintained with a volume-limited ventilator at 8- 10 breaths/ min, 15 cc/kg tidal volume, and 0.2 1 FIO2. After each animal was prepped and draped in a sterile fashion, a 12-cm left thoracotomy incision was performed at the fifth intercostal space. Prior to manipulation of internal organs, all instruments, gauze sponges, and gloves were bathed in the study solution. A 6-cm pericardiotomy was placed 3 cm anterior to the left phrenic nerve. This was immediately followed by irrigation of the pericardial sac with 20 cc of the study solution. The epicardium and the inner surface of the pericardium were abraded with 20 strokes of a gauze sponge to induce a punctated hemorrhagic lesion. The inner surface of the pericardium was irrigated three times with 20 cc of solution: immediately following pericardiotomy, 15 min following pericardiotomy, and prior to pericardial closure. After the epicardium and inner surface of the pericardium had been exposed to air for 30 min, to simulate intraoperative desiccation, the pericardium was closed loosely with 2-O silk sutures. Thoracotomy closure was performed in the usual fashion including aspiration of air and fluid from the left chest. Intercostal nerve blocks were induced with 1% xylocaine and l/ 100,000 epinephrine. Flocillin and Demerol were given for 7 postoperative days, and all dogs received regular diets. Reoperation for evaluation of pericardial adhesions was performed 6 weeks following
OF PERICARDIAL
ADHESIONS
45
initial surgery. Under general anesthesia all animals were subjected to bilateral thoracotomies. The pericardial cavity was evaluated for adhesions by two observers unaware of the study solution. A scale of O-4 was used to score adhesions of the abraded surface, unabraded surface, and suture line: O-no adhesions 1-filmy adhesions with easily identifiable plane 2-mild adhesions with freely dissectable plane 3-moderate adhesions with difficult dissection of plane 4-dense adhesions with nondissectable plane Photographs were taken of the pericardial cavity and adhesions (if present) to allow additional comparison and documentation (Figs. l-5). All animals were sacrificed and biopsies of abraded and unabraded epicardium and abraded and unabraded pericardium underwent histologic evaluation. The incidence and severity of adhesions in animals treated with hydrophilic polymer solutions (PVP and MCD) were compared with those of the control group treated with RL. Statistical analysis was performed with Fisher’s exact test. RESULTS
Pericardial adhesion scores are summarized in Table 1. There was no difference in
FIG. 1. PVP Grade 0. No adhesions were present at autopsy in this animal with polyvinylpyrrolidone solution.
46
JOURNAL
OF SURGICAL
RESEARCH:
VOL. 45, NO. 1, JULY 1988
FIG. 2. MCD Grade 0. No adhesions were present at autopsy in this animal irrigated with a methylcellulose derivative solution.
FIG. 4. RL Grade 4. Dense adhesions without dissectable planes were present at autopsy in this animal irrigated with Ringers lactate.
adhesion formation between the abraded surface and suture line. The RL-irrigated control group had adhesion scores >3 in all but one dog. None of these dogs were free of pericardial adhesions and the mean adhesion score was 3.2 + 1.1. Four of five PVP-treated dogs, on the contrary, showed no postoperative adhesion. The mean adhesion score was 0.2 + 0.4. An iatrogenic foreign body (sponge) was found in the lone PVP-treated animal who developed a mild pericardial adhesion. The MCD-treated group showed marked reduction of pericardial adhesion formation and had a mean adhesion score of 0.5 + 0.7. No animal in the MCD group developed a surgically significant adhesion (>2 adhesion score). A marked reduction in pericardial adhesion formation was shown when
comparing PVP and RL (P < 0.004) and when comparing MCD and RL (I’ < 0.024). No statistical difference existed between PVP and MCD (P < 0.5); however, there was a lower mean adhesion score with PVP (0.2) than with MCD (0.5). Histologic sections of epicardium and pericardium sent for evaluation of thickness, chronic inflammation, fibrosis, and neovasculasity revealed no trends or statistical differences in any parameter between PVP, MCD, and RL cohorts.
FIG. 3. RL Grade 2. Mild adhesions with freely dissectable planes were present at autopsy in this animal irrigated with Ringers lactate.
FIG. 5. RL Grade 4. Dense adhesions without dissectable planes were present at autopsy in this animal irrigated with Ringers lactate.
DISCUSSION
Incidence and Management of Pericardial Adhesions Cardiac reoperation constitutes 10% of heart surgery presently performed and its in-
DUNCAN
ET AL.: PREVENTION
OF PERICARDIAL
47
ADHESIONS
TABLE 1 POSTOPERATIVE PERICARDIAL ADHESIONS U
A
SL
C
RL,= RL2 RLS RL., RLS
0 0 0 0 0
4 2 3 2 4
4 I 4 4 4
4 1.5 3.5 3 4
Mean
0
3
3.4
3.2
U
A
SL
C
MCD, MCD2 MCDs MCD., MCDS
0 0 0 0 0
1 2 0 0 0
1 0 0 1 0
1 1 0 0.5 0
PVP, PVPZ PVPS PVP, PVPSh
0 0 0 0 0
0 0 0 0 1
0 0 0 0 1
0 0 0 0 1
Mean
0
0.6
0.4
0.5
Mean
0
0.2
0.2
0.2
U
Note. RL, Ringers lactate; MCD, methylcellulose derivative, PVP, polyvinylpyrrolidone; abraded; SL, suture line; C, combined abraded and suture line. a Sacrificed at 12 weeks because dog had puppies. b Two sponges inadvertently left in hemithorax.
A
SL
C
U, unabraded; A,
cidence continues to increase [ 151. The pri- instead of prevention. Dobel and Iain [2] mary indication for reoperation is coronary have outlined several steps for safer restemorevascularization [7]. Other frequent indicatomy: appreciate that a high risk situation tions include prosthetic valve failure and de- exists, review lateral chest radiographs, exfinitive cardiac repair following staged or pose the femoral artery and vein for potential palliative cardiac procedures in children with circulatory control with cardiopulmonary congenital heart disease. Following cardiac bypass, cautious reentry, retract sternal surgery, adhesions form between the epicar- tables upward, avoid premature placement dium and surrounding structures including of the sternal retractor, deflate lungs, and use the pericardium, mediastinal tissues, pleura, a low power oscillating saw [2, 71. Although and sternum. Reoperation is prolonged by surgical techniques for reoperation have imdifficult sternal reentry, adhesion lysis, and proved, resternotomy remains a challenge increased hemorrhage. Restemotomy is as- for the cardiac surgeon. sociated with a 2-6s incidence of major vascular injuries and a 37% mortality when major hemorrhage occurs [2, 71. Injuries to Adhesion Formation the innominate vein, superior vena cava, A variety of mechanisms have been shown right atrium, right ventricle, pulmonary arto be important in adhesion formation. All tery, and aorta are not uncommon. Aortogenerally involve the healing response to secoronary grafts involving the right coronary rosal injury. Pericardial injury occurs with artery and internal mammary arteries are desiccation, abrasion, and foreign body reacparticularly vulnerable to injury [7]. Furthertion. The etiology of surgical adhesion formore, adhesions and fibrosis obscure anatomic structures making identification of mation has been extensively reviewed [ 1, 4, 5,9, 14, 151. coronary vessels difficult. Five sequential steps seem to be involved: High risk situations for restemotomy include multiple reoperations, previous as- (1) initial injury to the surface endothelium, (2) pouring out of exudate at the site of incending aortic aneurysm repairs, multiple valve disease, right atrial or ventricular en- jury, (3) coagulation of the exudate and filargement, right ventricular-pulmonary ar- brin formation, (4) fibrin adhesiveness caustery conduits, and previous mediastinitis [7]. ing adjacent structures to stick together, and The present approach to pericardial adhe- (5) organization with fibrous tissue formasions involves attempts at safe restemotomy tion.
48
JOURNAL
OF
SURGICAL
RESEARCH:
VOL.
45, NO.
I, JULY
1988
Prevention of Adhesion Formation
Hydrophilic Polymer Solutions
Although the prevention of pericardial adhesions has only recently been investigated, much has been learned in the prevention of postoperative adhesions from research in abdominal and opthalmic surgery. Present efforts to prevent postoperative pericardial adhesions center on three principles: (1) preventing unnecessary serosal injury; (2) primary pericardial closure, pericardial augmentation, or pericardial substitutes; and (3) local or systemic pharmacology [ 1 I]. These efforts to date have reported minimal success. This study did not attempt to address the question of pericardial closure or substitutes. Although our study was confined to the pericardial cavity, an understanding of the mechanism of hydrophilic polymer solutions and their use in previous peritoneal studies suggest that the results may be extrapolated to extrapericardial structures. The relationship of serosal injury to postoperative adhesion formation has been documented. Crush injuries, abrasions, foreign body reactions, desiccation, infection, dry gloves, and sponges have all been implicated in adhesion formation. Primary pericardial closure had been advocated; however, it has largely been abandoned due to kinking of grafts and disturbances of cardiac filling [7]. Pericardial meshing has been reported to show some success [ 111. Although some pericardial substitutes appear to reduce adhesion formation, their use has been associated with severe epicardial fibrosis creating difficulty identifying coronary vessels [3, lo]. Systemic agents such as promethazine, dexamethazone, urokinase, streptokinase, heparin, and dextran have had Limited success; however, each is associated with significant side effects [ 15, 161. Local application of Dextran 70, in the rabbit pericardial cavity, to separate serosal surfaces, has reportedly shown some beneficial effect [ 141. Hydrophilic polymer solution tissue coatings have been proven to be effective in animal studies for the prevention of peritoneal adhesions, but have not been reported in pericardial adhesion studies to date [5, 91.
Studies on the reduction of surgical adhesions have focused on preventing the process of adhesion formation following significant tissue injury. An important aspect of this study was the use of protective nonadhesive tissue coatings with a hydrophilic polymer solution. Such tissue coatings may be expected to reduce desiccation and prevent sticking of tissues. The objective is to prevent tissue damage during surgical manipulation rather than reverse a surgical injury. Goldberg et al. showed a marked reduction of peritoneal adhesions with PVP solutions [5]. In addition, it was pointed out that significant serosal injuries are induced by manipulative abrasions due to sponges and surgeons’ gloves. Electron microscopy showed a major reduction in tissue damage following the application of PVP solutions [5]. SUMMARY
Our results indicate that hydrophilic polymer solutions such as PVP and MCD significantly reduce pericardial adhesion formation when compared to controls. Clinical application of hydrophilic polymer solutions for the prevention of pericardial adhesions should reduce morbidity and mortality of cardiac reoperation. Hydrophilic polymer solutions have been subjected to extensive toxicologic evaluations and have been found to have no adverse effects; however, additional research is needed to determine the most efficacious hydrophilic polymers, optimal molecular weights, concentrations, osmolarities, pH’s, and body clearances. Bioacceptance, long term effects, and methods of application for hydrophilic polymer solutions need further investigation. REFERENCES 1. ClifT, W. J., Grobety, J., and Ryan, G. B. Postoperative pericardial adhesions: The role of mild serosal injury and spilled blood. J. Thoruc. Curdiovasc. Surg. 65: 1973. 2. Dobell, R. C., and Iain, A. K. Catastrophic hemor-
DUNCAN
3.
4. 5.
6. 7. 8.
ET AL.:
PREVENTION
rhage during redo stemotomy. Ann. Thorac. Surg. 37: 1984. Gallo, I., Artinano, E., Duran, C. G. Late clinical results with the use of heterologous pericardium for closure of the pericardial cavity. J. Thorac. Cardiovast. Surg. 89: 109, 1985. Gervin, A. S., Jacobs, G., Hofnagel, H., and Mason, K. Surgical trauma and pericardial fibrinolytic activity. Amer. Surg. 225, 1915. Goldberg, E. P., Sheets, J. W., and Habal, M. B. Peritoneal adhesions: Prevention with the use of hydrophilic polymer coatings. Arch. Surg. 115: 776, 1980. Londe, S., and Sugg, W. L. The challenge of reoperation in cardiac surgery. Ann. Thorac. Surg. 17: 157, 1974. Loop, F. D. Catastrophic hemorrhage during sternal reentry. Ann. Thorac. Surg. 37: 271, 1984. Math&n, S. R., Hong-De, W., Savvage, L. R., and Walker, M. W. Prevention of retrostemal adhesions after pericardiotomy. J. Thorac. Cardiovasc. Surg. 92: 92, 1986.
Mazuji, M., and Fadhh, H. A. Peritoneal adhesions: Prevention with povidone and dextran 75. Arch. Surg. 91: 872, 1965. 10. Meus, P. J., Wemly, J. A., Campbell, C. D., Takanaski, Y., Pick, R. L., Qui Zhao-Kun, and Replogle, R. L. Long-term evaluation of pericardial substitutes. J. Thorac. Cardiovasc. Surg. 85: 54, 1983. 11. Milgaltes, E., Gideon, V., Siberman, S., Appelbaum, Y., Shimon, D., Kopolevic, J., Cohen, D., Jones, H., Appelbaum, A., and Berman, J. Pericar9.
OF
PERICARDIAL
ADHESIONS
49
dial meshing: An effective method for prevention of pericardial adhesions and epicardial reaction after cardiac operations. J. Thorac. Cardiovasc. Surg. 90: 281, 1985. 12. Oyer, P. E., and Shumway, N. E. Again, via the median stemotomy. Arch. Surg. 109: 604, 1974. 13. Pucci, J. J., Walesby, R. K., Smith, E. J., Edmondson, S. J., Sapsford, R. N., Bentall, H. H., and Cleland, W. P. Reoperation for failed aorto-coronary bypass grafts. J. Cardiovasc. Surg. 23: 453, 1982. 14. Robison, R. J., Brown, J. W., Deschnes, W. P., Highes, B., and King, H. Prevention of pericardial adhesions with Dextran 70. Ann. Thorac. Surg. 37: 488, 1984.
15. Salm, T. J., Okike, 0. N., Massicano, T. N., Comp ton, C., and Espinoza, E. Prevention of postoperative pericardial adhesions. Arch. Surg. 121: 462, 1986.
Smith, L. 0. Prevention of surgically induced pericardial adhesions with combined dexamethasone and promethazine therapy. J. Flu. Med. Assoc. 55: 413, 1968. 17. Ungerlieder, R. M., Mills, N. L., and Wechsler, A. S. Left thoracotomy for reoperative coronary artery bypass procedures. Ann. Thorac. Surg. 40: 11, 1985. 18. Urschel, H. C., Razzok, M. A., and Gardner, M. Coronary artery bypass occlusion secondary to postcardiotomy syndrome. Ann. Thorac. Surg. 22: 528, 1976. 19. Youmans, C. R., White, J., and Derrick, J. R. The prevention of pleural and pericardial adhesions with silastic. J. Thorac. Cardiovasc. Surg. 55: 383, 1968. 16.