- Email: [email protected]
Percutaneous abdominal abscess drainage
Percutaneous Abdominal Abscess Drainage Portland Area Experience
Jeffrey Sunshine, MD, Portland, Oregon Donald B. McConnell, MD, Portland, Oregon Carol J. Weinstein, MD, Portland, Oregon Truman M. Sasaki, MD, Portland, Oregon R. Mark Vetto, MD, Portland, Oregon
Recent refinements in imaging technology have been combined with sophisticated radiologic techniques [1,2] to produce a series of published reports that describe successful drainage of intraabdominal abscess with percutaneously inserted catheters. Using real-time ultrasonography, rapid (2 second) computerized tomographic scanning, and fluoroscopy, an impressive 73 to 86 percent of intraabdominal abscesses were recently reported as definitively drained without need of a general anesthetic or laparotomy [3-6]. Because of the historically high rates on morbidity and mortality attendant to the surgical management of intraabdominal abscesses, whether intraabdominal, retroperitoneal, or visceral [7-IO], this reported experience certainly warrants careful scrutiny and further validation where possible. For this reason, we have retrospectively reviewed the 3 year experience in six Portland area hospitals with percutaneous abdominal abscess drainage in an effort to further assess the efficacy of the procedure. Material
and Methods
A total of 21 patients from six Portland metropolitan area hospitals were reviewed (Oregon Health Sciences University, Portland Veterans Administration, Good Samaritan Hospital, St. Vincent Hospital, Emanuel Hospital, and Bess Kaiser Hospital). Each hospital contributed approximately three patients (range, one to eight). All paFrom the Department of Surgery, School of Medicine. University of Oregon Health Sciences University, and the Surgical Service, Veterans Administration Medical Center, Portland, Oregon. Requests for reprints should be addressed to Donald B. McConnell, MD, Surgical Service (112P), Portland Veterans Administration Medical Center, P.O. Box 1034, Portland, Oregon 97207. Presented at the 69th Annual Meeting of the North Pacific Surgical Association, Spokane, Washington, November 12 and 13, 1982.
Volume 145, May 1983
tients had previously undergone at least one abdominal surgical procedure (some as many as three) before being referred for percutaneous abdominal abscess drainage. A patient was usually referred to the hospital’s radiology department on the basis of postoperative suspicion of an abscess often accompanied by fever and leukocytosis, with or without a palpable mass. Ultrasonography or, in a few patients, computerized tomography was then used as the initial imaging technique which revealed an intraabdominal space-occupying mass (by abdominal we mean visceral and retroperitoneal, as well as intraabdominal, unless otherwise specified). When the decision was made that the intraabdominal mass needed drainage but that the patient at the time was best approached nonsurgically, percutaneous abdominal abscess drainage was carried out. Although all radiology departments had access to fluoroscopes, ultrasonographic devices, and computerized tomographic scanners, the manufacturers of these instruments, as well as the “newness” or “generation” of computerized tomographic scanners and ultrasound devices, varied. Personal preference then dictated in which order or combination the imaging devices were used for planning the access and puncture of the intraabdominal mass. For example, ultrasonography may have been used to localize a mass, but to avoid inadvertent sampling of bowel contents and to disclose a second or synchronous mass more clearly, a computerized tomographic scan may have been performed as well. The latter scan could then help define the safest approach to the pathologic abnormality. The radiologist may then choose to use a rapid 2 second computerized tomographic scanner, real-time ultrasonography, or fluoroscopy to puncture and drain the mass. Fluoroscopy was especially useful in visualizing the cavity with needle and contrast medium present. The type of cannulation system employed varied. Commonly a 22 gauge Chiba needle was inserted percutaneously into the intraabdominal mass. A variable amount of aspirated contents was sent for Gram stain, culture and
615
Sunshine et al
TABLE I
Successful Dralnage Procedures
Age W
Drainage
and Sex
Site and Date
37, M 26, M 61, M 81, M 40, M 59, F 50, M 79, F 55, M
Lesser sac, LUQ 10/g/81 7 cm head of pancreas 1 l/79’ Right intrahepatic l/12/82 Right lobe of liver l/19/79 8 cm pelvic 12/19/81 Right intrahepatic 4/18/80 Left flank, unilocular 4/ 14/8 1 Left subphrenic 2110182 Right biliary l/28/81
Follow-up
Date and Type of Preabscess Surgery
(mo) 5
. 1 22 3
Total pancreatectomy 1976, strangulation small bowel obstruction, ileostomy, Hartman procedure 10/2/81 Splenectomy 5/6/79, open left subphrenic abscess 5/17/79, pancreatic abscess 6/l l/79, subphrenic abscess 6118179 Cholecystectomy. gastrojejunostomy, attempted total pancreatectomy 12/4/79 Choledochoduodenostomy (stones) 1971, open drainage of right hepatic abscess 12178 Perforated diverticulitis, no surgery
23
Cholecystectomy
11
Aortobifemoral l/12/81, right colectomy l/30/81, revised ileostomy, no abscess Omental patch, perforated duodenal ulcer l/29/82
left colectomy
Total gastrectomy, 1 l/9/79
colotomy, splenectomy
1 2
Recurred at 11 months. LUQ = left upper quadrant; PAAD = percutaneous
418180
Roux-Y, esophagojejunostomy,
2/2/81,
PAAD,
l
abdominal abscess drainage.
sensitivity reports, and polymorphonuclear cell count and differential. A guide wire was then placed through the Chiba needle into the abscess cavity and the Chiba needle was removed. This was followed by placement of increasing sizes of French dilators over the guide wire into the abscess. When a sufficient lumen into the abscess was attained, usually with a no. 7 or 8 F. dilator, a comparable pigtail catheter was placed over the guide wire in a fashion that caused it to curl within the abscess cavity. The guide wire was then removed and the soft pigtail catheter with multiple side holes was left within the cavity and then sutured
to the outer abdominal wall. This procedure is known as the modified Seldinger technique and is described in more detail elsewhere [2]. After placement of the pigtail catheter within the abscess cavity, the next decision was when or how often to irrigate the catheter and when to remove the catheter definitively. Like traditional open surgical techniques of incision and drainage, this aspect of care depended on the diminution of drainage and the clinical well-being of the patient, as well as on radiographic evidence of collapse of the abscess cavity. Catheters were sometimes removed as soon as initial drainage was completed, or else the catheter was left in place for several days. In one successful case it remained in place for 24 days. Because of the tendency of abscess cavities to collapse and subsequently expel the catheter, frequent checks were made to assess the catheter position; it was not uncommon for one or two draining side holes to be outside of the abscess cavity. Results
Successful drainage involved resolution of the abdominal process as evidenced by computerized tomography, ultrasonography, or fluoroscopy. In addition, we regarded a successful procedure as one in which the abscess was absent as subsequently confirmed at surgery, despite the persistence of a fever. Our rationale was that a definitive surgical
616
drainage is sometimes accompanied by a continued septic course and that a percutaneous attempt at abscess drainage should not be used to do more than the definitive surgical procedure. We also defined percutaneous abdominal abscess drainage as successful when the abscess did not recur for at least 1 month after removal of the catheter, with accompanying improvement in the clinical well-being of the patient as evidenced by defervescence and normalizing polymorphonuclear leukocyte counts. In this regard, 9 of 21 abscesses (43 percent) were successfully drained (Table I). We also evaluated major complications associated with this procedure and defined them as follows: inducing a decrease in the hematocrit value of greater than 10 percent, inducing sepsis as manifest by spiking fever with concomitant blood pressure reduction below 90 mm Hg, and producing respiratory or other distant organ failure within 12 hours of the procedure. According to these criteria, 5 of 21 procedures produced a major complication (24 percent). We regarded a minor complication as the inability to enter the abscess cavity or, if entered, an inability to drain the abscess. Problems with catheters plugging or becoming dislodged were also regarded as minor complications. There were a total of 13 minor complications in 21 patients (minor complication rate 61 percent). Finally, our review focused on the need for follow-up surgery. We tabulated whether reoperative surgery within 14 days of the catheterization attempt was required for major complications, continued symptoms, or both. If on reexploration no abscess was encountered, we still viewed surgery as required because no patient was reexplored who did not have persistent symptoms. We found that 8 of 21 patients
The Amerlcan Journal of Surgery
Percutaneous Abdominal Abscess Drainage
who underwent percutaneous abdominal abscess drainage ultimately required surgery to definitively treat or further characterize their abnormalities (reoperation rate 38 percent). Comments Our review of the Portland area experience with the percutaneous abdominal abscess drainage procedure did not corroborate the high success rates of four major previous studies of the procedure [3-61, In those studies, successful drainage occurred in 86, 86,73, and 85 percent of patients versus 43 percent of patients in the present review. However, our review was retrospective in nature, utilizing a number of institutions and, as such, a variety of personnel. The largest number of successfully drained abscesses occurred within the hepatobiliary system (four patients) after such diverse operations as cholecystectomy, choledochoduodenostomy for stones, and total gastrectomy with Roux-Y esophagojejunostomy for gastric carcinoma. In each patient percutaneous abdominal abscess drainage was attempted as a preliminary step before definitive drainage was carried out; however, it was successful and avoided major laparotomy in all of the patients. The presence of synchronous abscesses was ruled out by computerized tomographic scan, hence the fear of leaving an abscess undrained was diminished. At present, of the eight clearly successful drainages from all sites, five of the patients underwent the procedure less than 6 months ago. One successfully drained abscess (by our criteria) required surgical redrainage 11 months later. As described in previously published reports, the abscesses most successfully drained were those that were unilocular without septations. Infected hematomas containing nonliquid necrotic material and other detritus as well as highly viscous purulence did not drain well, if at all. Phlegmons and multiple or miliary abscesses cannot be evacuated or definitively treated using these techniques. Because of the considerable major complication rate in this series (24 percent), we believe it incumbent that our radiologic colleagues characterize these intraabdominal processes as fully as possible before the procedure is performed in order to effect the best results with the least morbidity. In addition, we feel strongly that only “postsurgical abscesses” as opposed to “spontaneous abscesses” should be considered for a percutaneous abdominal abscess drainage procedure. In postsurgical settings an understanding of the original pathologic condition is usually known and visualized. The risk of draining an abscess emanating from a necrotic gallbladder, perforated viscus, or ruptured appendix, without correcting the underlying source of infection, is too high to justify such a procedure. We believe a surgeon should be apprised of the timing of the procedure so that operating room time may be set aside in case it is needed emergently.
Volume 145, May 1993
Despite a comparatively lower percentage of successfully drained abscesses and a comparatively higher percentage of complications, 43 percent of patients were spared the use of a general anesthetic and exploratory laparotomy because of this procedure. Because the principles of drain management are comparable in percutaneously and intraoperatively placed drains, it is not surprising that the duration of catheterization for successfully drained abscesses ranged from 7 to 24 days. It is our opinion that successfully drained abscesses were those that were irrigated frequently, always with high levels of antibiotics present in the blood during irrigation, because sepsis constituted the biggest risk in these procedures. Irrigation of the catheter at least every 4 to 6 hours helped prevent clogging. In terms of major complications, intrahepatic abscesses led the list (three of five complications). They follow: An 81 year old man who had undergone a choledochoduodenostomy for stone disease in 1971 had a right intrahepatic abscess drained surgically 7 years later. His abscess recurred within 1 month of drainage and he was referred for percutaneous abdominal abscess drainage and was treated successfully by this technique. However, within hours after a percutaneous catheter was placed he became diaphoretic with chills. An orthostatic blood pressure drop occurred, and he became apneic. Aspirate grew Proteus, Klebsiella, and Streptococcus fecalis, but the patient survived the acute setback without recurrent abscess for 22 months. A 33 year old woman with stage IV non-Hodgkin’s lymphoma presented with a fluid-filled cavity within the liver anterior to the portal vein. She had undergone right hemicolectomy 4 months earlier for bowel infarction and extensive retroperitoneal lymphoma. No purulence was returned from the initial aspirate, only clotted blood which subsequently grew Bacteriodes fragilis, Escherichia coli, and Strep. facalis. The evening after aspiration her hematocrit value dropped from 35 to 27 percent and her systolic blood pressure dropped from 130 to 80 mm Hg. Pneumonia of the right middle lobe developed 3 days later. She seized and died the following day. Because the patient had fever before the procedure, her rapidly progressive downhill course could not be causally linked to the percutaneous abdominal abscess drainage procedure. This case underscores the fact that septic and hemorrhagic complications can occur with percutaneous abdominal abscess drainage and that infected hematomas cannot be drained with these techniques. A third complication involved a 69 year old man in whom pancreatitis in the absence of alcohol abuse, gallstones, ulcers, or thiazide diuretics developed. After stabilizing clinically, this patient was readmitted, and a mass at the head and uncinate process of the pancreas was approached percutaneously with the presumptive diagnosis of enlarging pancreatic
617
Sunshine
et al
carcinoma (the mass was enlarged on serial ultrasonograms but his arteriogram showed no abnormalities). The procedure induced a 5 day course of pancreatitis which later resolved. However, an aspirate sent for cytologic examination suggested the presence of a possible liposarcoma. Exploratory laparotomy was undertaken shortly thereafter which revealed extensive fat necrosis with no definite diagnosis of the mass. This case illustrates the point that attempts at definitive management of “spontaneous” fluid collections should not be undertaken with this technique, because the primary abnormality can be obfuscated, and in this case has remained unknown. Summary After reviewing 21 patients who have had percutaneous abdominal abscess drainage, we believe that the procedure should be considered for those abscesses that are unilocular without septations, with safe access being a key variable dictating the use of percutaneous abdominal abscess drainage rather than surgery. A computerized tomographic scan of the abdomen should be employed at some stage of the percutaneous abdominal abscess drainage procedure to facilitate safe access to the abscess and to distinguish a synchronous abscess where present. In addition, we believe that percutaneous abdominal abscess drainage should be considered for postsurgical abscesses only and not those that are spontaneous in nature or where the original abnormality cannot be accurately surmised. With regard to catheter management, frequent irrigation of the catheter must be carried out at least every 4 to 6 hours, with high levels of antibiotics present in the blood before irrigation. This must be done to obviate
618
the most frequent and potentially lethal complication of the procedure, namely sepsis. Percutaneous abdominal abscess drainage, although safe for the most part, is capable of inducing considerable morbidity. Our data suggest that percutaneous abdominal abscess drainage is not as efficacious as previous reports have suggested. Traditional surgical drainage techniques are best utilized for those abscesses that are multiple, highly viscous, inaccessible, spontaneous, or unresponsive to percutaneous abdominal abscess drainage. References 1. Gerzoff S. Triangulation: indirect CT guidance for abscess drainage. AJR 1981;137:1080-1. 2. Haaga JR, Lipuma JP, Eckhauser ML. A piggyback technique for percutaneous insertion of large catheters. AJR 1981; 138:1245-8. 3. Gerzoff S, Robbins AH, Johnson WC, Birkett DH, Nabseth DC. Percutaneous catheter drainage of abdominal abscesses--A five year experience. N Engi J Med 1981;305:653-7. 4. Van Sonnenberg E, Ferrvei JT, Mueller PR, Wittenberg J, Simeone JF. Percutaneous drainage of abscesses and fluid collections: technique, results, and applications. Radiology 1982;142:1-10. 5. Martin EC, Karlson KB, Fankuchen El, Cooperman A, Casarella WJ. Percutaneous drainage of postoperative intraabdominal abscesses. AJR 1982;138:13-5. 6. Haaga JR, Weinstein JA. CT guided percutaneous aspiration and drainage of abscesses. AJR 1980; 135: 1187-94. 7. Wang SM, Wilson SE. Subphrenic abscess: the new epidem iology. Arch Surg 1977;112:934. 8. Sherman NJ, Davis JR. Subphrenic abscess: a continuing hazard. Am J Surg 1969;117:117. 9. Altemeir WA, Culbertson WR. lntraabdominal abscess. Am J Surg 1973;125:70. 10. Welch CE, Nedberg SE. Suppurative peritonitis with major abscesses. In: Artz CD, Hardy JD, eds. Complications in surgery and their management. Philadelphia: WB Saunders, 1967:577.
The American Journal of Surgery
Jeffrey Sunshine, MD, Portland, Oregon Donald B. McConnell, MD, Portland, Oregon Carol J. Weinstein, MD, Portland, Oregon Truman M. Sasaki, MD, Portland, Oregon R. Mark Vetto, MD, Portland, Oregon
Recent refinements in imaging technology have been combined with sophisticated radiologic techniques [1,2] to produce a series of published reports that describe successful drainage of intraabdominal abscess with percutaneously inserted catheters. Using real-time ultrasonography, rapid (2 second) computerized tomographic scanning, and fluoroscopy, an impressive 73 to 86 percent of intraabdominal abscesses were recently reported as definitively drained without need of a general anesthetic or laparotomy [3-6]. Because of the historically high rates on morbidity and mortality attendant to the surgical management of intraabdominal abscesses, whether intraabdominal, retroperitoneal, or visceral [7-IO], this reported experience certainly warrants careful scrutiny and further validation where possible. For this reason, we have retrospectively reviewed the 3 year experience in six Portland area hospitals with percutaneous abdominal abscess drainage in an effort to further assess the efficacy of the procedure. Material
and Methods
A total of 21 patients from six Portland metropolitan area hospitals were reviewed (Oregon Health Sciences University, Portland Veterans Administration, Good Samaritan Hospital, St. Vincent Hospital, Emanuel Hospital, and Bess Kaiser Hospital). Each hospital contributed approximately three patients (range, one to eight). All paFrom the Department of Surgery, School of Medicine. University of Oregon Health Sciences University, and the Surgical Service, Veterans Administration Medical Center, Portland, Oregon. Requests for reprints should be addressed to Donald B. McConnell, MD, Surgical Service (112P), Portland Veterans Administration Medical Center, P.O. Box 1034, Portland, Oregon 97207. Presented at the 69th Annual Meeting of the North Pacific Surgical Association, Spokane, Washington, November 12 and 13, 1982.
Volume 145, May 1983
tients had previously undergone at least one abdominal surgical procedure (some as many as three) before being referred for percutaneous abdominal abscess drainage. A patient was usually referred to the hospital’s radiology department on the basis of postoperative suspicion of an abscess often accompanied by fever and leukocytosis, with or without a palpable mass. Ultrasonography or, in a few patients, computerized tomography was then used as the initial imaging technique which revealed an intraabdominal space-occupying mass (by abdominal we mean visceral and retroperitoneal, as well as intraabdominal, unless otherwise specified). When the decision was made that the intraabdominal mass needed drainage but that the patient at the time was best approached nonsurgically, percutaneous abdominal abscess drainage was carried out. Although all radiology departments had access to fluoroscopes, ultrasonographic devices, and computerized tomographic scanners, the manufacturers of these instruments, as well as the “newness” or “generation” of computerized tomographic scanners and ultrasound devices, varied. Personal preference then dictated in which order or combination the imaging devices were used for planning the access and puncture of the intraabdominal mass. For example, ultrasonography may have been used to localize a mass, but to avoid inadvertent sampling of bowel contents and to disclose a second or synchronous mass more clearly, a computerized tomographic scan may have been performed as well. The latter scan could then help define the safest approach to the pathologic abnormality. The radiologist may then choose to use a rapid 2 second computerized tomographic scanner, real-time ultrasonography, or fluoroscopy to puncture and drain the mass. Fluoroscopy was especially useful in visualizing the cavity with needle and contrast medium present. The type of cannulation system employed varied. Commonly a 22 gauge Chiba needle was inserted percutaneously into the intraabdominal mass. A variable amount of aspirated contents was sent for Gram stain, culture and
615
Sunshine et al
TABLE I
Successful Dralnage Procedures
Age W
Drainage
and Sex
Site and Date
37, M 26, M 61, M 81, M 40, M 59, F 50, M 79, F 55, M
Lesser sac, LUQ 10/g/81 7 cm head of pancreas 1 l/79’ Right intrahepatic l/12/82 Right lobe of liver l/19/79 8 cm pelvic 12/19/81 Right intrahepatic 4/18/80 Left flank, unilocular 4/ 14/8 1 Left subphrenic 2110182 Right biliary l/28/81
Follow-up
Date and Type of Preabscess Surgery
(mo) 5
. 1 22 3
Total pancreatectomy 1976, strangulation small bowel obstruction, ileostomy, Hartman procedure 10/2/81 Splenectomy 5/6/79, open left subphrenic abscess 5/17/79, pancreatic abscess 6/l l/79, subphrenic abscess 6118179 Cholecystectomy. gastrojejunostomy, attempted total pancreatectomy 12/4/79 Choledochoduodenostomy (stones) 1971, open drainage of right hepatic abscess 12178 Perforated diverticulitis, no surgery
23
Cholecystectomy
11
Aortobifemoral l/12/81, right colectomy l/30/81, revised ileostomy, no abscess Omental patch, perforated duodenal ulcer l/29/82
left colectomy
Total gastrectomy, 1 l/9/79
colotomy, splenectomy
1 2
Recurred at 11 months. LUQ = left upper quadrant; PAAD = percutaneous
418180
Roux-Y, esophagojejunostomy,
2/2/81,
PAAD,
l
abdominal abscess drainage.
sensitivity reports, and polymorphonuclear cell count and differential. A guide wire was then placed through the Chiba needle into the abscess cavity and the Chiba needle was removed. This was followed by placement of increasing sizes of French dilators over the guide wire into the abscess. When a sufficient lumen into the abscess was attained, usually with a no. 7 or 8 F. dilator, a comparable pigtail catheter was placed over the guide wire in a fashion that caused it to curl within the abscess cavity. The guide wire was then removed and the soft pigtail catheter with multiple side holes was left within the cavity and then sutured
to the outer abdominal wall. This procedure is known as the modified Seldinger technique and is described in more detail elsewhere [2]. After placement of the pigtail catheter within the abscess cavity, the next decision was when or how often to irrigate the catheter and when to remove the catheter definitively. Like traditional open surgical techniques of incision and drainage, this aspect of care depended on the diminution of drainage and the clinical well-being of the patient, as well as on radiographic evidence of collapse of the abscess cavity. Catheters were sometimes removed as soon as initial drainage was completed, or else the catheter was left in place for several days. In one successful case it remained in place for 24 days. Because of the tendency of abscess cavities to collapse and subsequently expel the catheter, frequent checks were made to assess the catheter position; it was not uncommon for one or two draining side holes to be outside of the abscess cavity. Results
Successful drainage involved resolution of the abdominal process as evidenced by computerized tomography, ultrasonography, or fluoroscopy. In addition, we regarded a successful procedure as one in which the abscess was absent as subsequently confirmed at surgery, despite the persistence of a fever. Our rationale was that a definitive surgical
616
drainage is sometimes accompanied by a continued septic course and that a percutaneous attempt at abscess drainage should not be used to do more than the definitive surgical procedure. We also defined percutaneous abdominal abscess drainage as successful when the abscess did not recur for at least 1 month after removal of the catheter, with accompanying improvement in the clinical well-being of the patient as evidenced by defervescence and normalizing polymorphonuclear leukocyte counts. In this regard, 9 of 21 abscesses (43 percent) were successfully drained (Table I). We also evaluated major complications associated with this procedure and defined them as follows: inducing a decrease in the hematocrit value of greater than 10 percent, inducing sepsis as manifest by spiking fever with concomitant blood pressure reduction below 90 mm Hg, and producing respiratory or other distant organ failure within 12 hours of the procedure. According to these criteria, 5 of 21 procedures produced a major complication (24 percent). We regarded a minor complication as the inability to enter the abscess cavity or, if entered, an inability to drain the abscess. Problems with catheters plugging or becoming dislodged were also regarded as minor complications. There were a total of 13 minor complications in 21 patients (minor complication rate 61 percent). Finally, our review focused on the need for follow-up surgery. We tabulated whether reoperative surgery within 14 days of the catheterization attempt was required for major complications, continued symptoms, or both. If on reexploration no abscess was encountered, we still viewed surgery as required because no patient was reexplored who did not have persistent symptoms. We found that 8 of 21 patients
The Amerlcan Journal of Surgery
Percutaneous Abdominal Abscess Drainage
who underwent percutaneous abdominal abscess drainage ultimately required surgery to definitively treat or further characterize their abnormalities (reoperation rate 38 percent). Comments Our review of the Portland area experience with the percutaneous abdominal abscess drainage procedure did not corroborate the high success rates of four major previous studies of the procedure [3-61, In those studies, successful drainage occurred in 86, 86,73, and 85 percent of patients versus 43 percent of patients in the present review. However, our review was retrospective in nature, utilizing a number of institutions and, as such, a variety of personnel. The largest number of successfully drained abscesses occurred within the hepatobiliary system (four patients) after such diverse operations as cholecystectomy, choledochoduodenostomy for stones, and total gastrectomy with Roux-Y esophagojejunostomy for gastric carcinoma. In each patient percutaneous abdominal abscess drainage was attempted as a preliminary step before definitive drainage was carried out; however, it was successful and avoided major laparotomy in all of the patients. The presence of synchronous abscesses was ruled out by computerized tomographic scan, hence the fear of leaving an abscess undrained was diminished. At present, of the eight clearly successful drainages from all sites, five of the patients underwent the procedure less than 6 months ago. One successfully drained abscess (by our criteria) required surgical redrainage 11 months later. As described in previously published reports, the abscesses most successfully drained were those that were unilocular without septations. Infected hematomas containing nonliquid necrotic material and other detritus as well as highly viscous purulence did not drain well, if at all. Phlegmons and multiple or miliary abscesses cannot be evacuated or definitively treated using these techniques. Because of the considerable major complication rate in this series (24 percent), we believe it incumbent that our radiologic colleagues characterize these intraabdominal processes as fully as possible before the procedure is performed in order to effect the best results with the least morbidity. In addition, we feel strongly that only “postsurgical abscesses” as opposed to “spontaneous abscesses” should be considered for a percutaneous abdominal abscess drainage procedure. In postsurgical settings an understanding of the original pathologic condition is usually known and visualized. The risk of draining an abscess emanating from a necrotic gallbladder, perforated viscus, or ruptured appendix, without correcting the underlying source of infection, is too high to justify such a procedure. We believe a surgeon should be apprised of the timing of the procedure so that operating room time may be set aside in case it is needed emergently.
Volume 145, May 1993
Despite a comparatively lower percentage of successfully drained abscesses and a comparatively higher percentage of complications, 43 percent of patients were spared the use of a general anesthetic and exploratory laparotomy because of this procedure. Because the principles of drain management are comparable in percutaneously and intraoperatively placed drains, it is not surprising that the duration of catheterization for successfully drained abscesses ranged from 7 to 24 days. It is our opinion that successfully drained abscesses were those that were irrigated frequently, always with high levels of antibiotics present in the blood during irrigation, because sepsis constituted the biggest risk in these procedures. Irrigation of the catheter at least every 4 to 6 hours helped prevent clogging. In terms of major complications, intrahepatic abscesses led the list (three of five complications). They follow: An 81 year old man who had undergone a choledochoduodenostomy for stone disease in 1971 had a right intrahepatic abscess drained surgically 7 years later. His abscess recurred within 1 month of drainage and he was referred for percutaneous abdominal abscess drainage and was treated successfully by this technique. However, within hours after a percutaneous catheter was placed he became diaphoretic with chills. An orthostatic blood pressure drop occurred, and he became apneic. Aspirate grew Proteus, Klebsiella, and Streptococcus fecalis, but the patient survived the acute setback without recurrent abscess for 22 months. A 33 year old woman with stage IV non-Hodgkin’s lymphoma presented with a fluid-filled cavity within the liver anterior to the portal vein. She had undergone right hemicolectomy 4 months earlier for bowel infarction and extensive retroperitoneal lymphoma. No purulence was returned from the initial aspirate, only clotted blood which subsequently grew Bacteriodes fragilis, Escherichia coli, and Strep. facalis. The evening after aspiration her hematocrit value dropped from 35 to 27 percent and her systolic blood pressure dropped from 130 to 80 mm Hg. Pneumonia of the right middle lobe developed 3 days later. She seized and died the following day. Because the patient had fever before the procedure, her rapidly progressive downhill course could not be causally linked to the percutaneous abdominal abscess drainage procedure. This case underscores the fact that septic and hemorrhagic complications can occur with percutaneous abdominal abscess drainage and that infected hematomas cannot be drained with these techniques. A third complication involved a 69 year old man in whom pancreatitis in the absence of alcohol abuse, gallstones, ulcers, or thiazide diuretics developed. After stabilizing clinically, this patient was readmitted, and a mass at the head and uncinate process of the pancreas was approached percutaneously with the presumptive diagnosis of enlarging pancreatic
617
Sunshine
et al
carcinoma (the mass was enlarged on serial ultrasonograms but his arteriogram showed no abnormalities). The procedure induced a 5 day course of pancreatitis which later resolved. However, an aspirate sent for cytologic examination suggested the presence of a possible liposarcoma. Exploratory laparotomy was undertaken shortly thereafter which revealed extensive fat necrosis with no definite diagnosis of the mass. This case illustrates the point that attempts at definitive management of “spontaneous” fluid collections should not be undertaken with this technique, because the primary abnormality can be obfuscated, and in this case has remained unknown. Summary After reviewing 21 patients who have had percutaneous abdominal abscess drainage, we believe that the procedure should be considered for those abscesses that are unilocular without septations, with safe access being a key variable dictating the use of percutaneous abdominal abscess drainage rather than surgery. A computerized tomographic scan of the abdomen should be employed at some stage of the percutaneous abdominal abscess drainage procedure to facilitate safe access to the abscess and to distinguish a synchronous abscess where present. In addition, we believe that percutaneous abdominal abscess drainage should be considered for postsurgical abscesses only and not those that are spontaneous in nature or where the original abnormality cannot be accurately surmised. With regard to catheter management, frequent irrigation of the catheter must be carried out at least every 4 to 6 hours, with high levels of antibiotics present in the blood before irrigation. This must be done to obviate
618
the most frequent and potentially lethal complication of the procedure, namely sepsis. Percutaneous abdominal abscess drainage, although safe for the most part, is capable of inducing considerable morbidity. Our data suggest that percutaneous abdominal abscess drainage is not as efficacious as previous reports have suggested. Traditional surgical drainage techniques are best utilized for those abscesses that are multiple, highly viscous, inaccessible, spontaneous, or unresponsive to percutaneous abdominal abscess drainage. References 1. Gerzoff S. Triangulation: indirect CT guidance for abscess drainage. AJR 1981;137:1080-1. 2. Haaga JR, Lipuma JP, Eckhauser ML. A piggyback technique for percutaneous insertion of large catheters. AJR 1981; 138:1245-8. 3. Gerzoff S, Robbins AH, Johnson WC, Birkett DH, Nabseth DC. Percutaneous catheter drainage of abdominal abscesses--A five year experience. N Engi J Med 1981;305:653-7. 4. Van Sonnenberg E, Ferrvei JT, Mueller PR, Wittenberg J, Simeone JF. Percutaneous drainage of abscesses and fluid collections: technique, results, and applications. Radiology 1982;142:1-10. 5. Martin EC, Karlson KB, Fankuchen El, Cooperman A, Casarella WJ. Percutaneous drainage of postoperative intraabdominal abscesses. AJR 1982;138:13-5. 6. Haaga JR, Weinstein JA. CT guided percutaneous aspiration and drainage of abscesses. AJR 1980; 135: 1187-94. 7. Wang SM, Wilson SE. Subphrenic abscess: the new epidem iology. Arch Surg 1977;112:934. 8. Sherman NJ, Davis JR. Subphrenic abscess: a continuing hazard. Am J Surg 1969;117:117. 9. Altemeir WA, Culbertson WR. lntraabdominal abscess. Am J Surg 1973;125:70. 10. Welch CE, Nedberg SE. Suppurative peritonitis with major abscesses. In: Artz CD, Hardy JD, eds. Complications in surgery and their management. Philadelphia: WB Saunders, 1967:577.
The American Journal of Surgery