The Evolution of Image-Guided Percutaneous Abscess Drainage: A Short History

What is the definition of “greatness” in intervention? Success? Longevity and success? Longevity, success, and few complications? Longevity, success, few complications, and a procedure which is almost exclusively performed by radiologists and unlikely to be “taken over” by other specialties? I am not entirely sure, but, over its 25-year history, percutaneous abscess drainage (PAD) has evolved from a specialized technique performed on the most obvious and simple of collections, to the best and most efficacious of all interventional procedures performed on almost any fluid collection in almost any location.

The concept of “minimally invasive drainage” did not start with radiology. In fact surgeons such as Debakey and Welch defined these terms in the golden years of abdominal surgery in the 1950s, 60s, and 70s. Clearly, many surgical articles described “minimally invasive” approaches to subphrenic and other abscesses. However, if there is one procedure which exemplified successful minimally invasive therapy, it is PAD.

Perhaps the first percutaneous use of needle aspiration and/or drainage was in the mid-1950s. McFadzen,[1] in fact, in the British Journal of Surgery in 1954, described “aspiration of liver abscesses with needle” as a curative procedure. The imaging of such lesions in 1954 may have been difficult, but aspiration of the majority of the purulent material combined with antibiotics certainly worked. Still, his group met with a negative reaction and his work was challenged in the literature. Ogden[2] and associates in 1961 apparently wrote that this aspiration procedure could cause peritoneal and pleural contamination. Surprisingly, these concerns still exist and many individuals have stated in workshops I have attended that they would defer performing a subphrenic drainage because of the potential of contaminating the pleural space.

Because of lack of imaging, not much evolved with percutaneous drainage until the mid-1970s. However, Pedersen[3] and his group did report the use of ultrasound to drain a “renal carbuncle” in 1973. Hans Holm[3] and his colleagues from Copenhagen were also instrumental in describing the use of ultrasound in guiding punctures of a variety of structures for biopsy and aspiration. Hans was a gregarious, English-speaking radiologist who visited the United States in the late 70s and early 80s to lecture and educate radiologists on this wonderful use of ultrasound. In addition to Holm,[3] Smith[4] and Bartrum reported the use of ultrasound guidance for abscess drainage in 1974 in the American Journal of Roentgenology (AJR). Doust[5] and colleagues described the use of ultrasound drainage for abdominal fluid collections in radiology in 1977. Gronvall[6] and associates, a European group, followed these early reports with a paper in AJR in 1977. Clearly, the Europeans, who were very familiar with ultrasound, were “early in the game” in using imaging for PAD.

It is easier to document the history of abscess drainage in the United States than in Europe or the Far East, but obviously simultaneous work was going on. Tetz[7] and associates reported the use of “isotope scan localization” to guide the drainage of liver abscesses as far back as 1973.

Simultaneous advancements in vascular and nonvascular radiology aided the use of instruments for PAD. The Seldinger technique which so revolutionized vascular work was first reported in the 1970s. This description enabled subsequent pioneers to actually place catheters rather than needles into collections.

Because ultrasound was more universally popular and available, many of the early reports were of simple aspiration of fluid collections using this device. In addition to Holm and Pederson, Goldberg[8] and Goldman[9] described the early use of ultrasound for simple fluid aspiration.

However, because of the many limitations of ultrasound visualization, the real advances in abscess drainage were due to the anatomical visualization afforded by CT. The limitations of ultrasound are obvious, but bear repeating because most of us who started draining abscesses in the 1970s had to understand them, and they still are important today.

Generally, ultrasound drainage is possible and successful for (1) superficially located collections; (2) larger collections; and (3) collections in parenchymal organs. In the later 1970s and early 1980s the ultrasound equipment was more primitive. Now, in 2003, collections as small as 2 cm that are interloop in location may be aspirated, whereas 20 years ago they probably were not seen. Unilocular collections are more likely to be completely drained, although this too is true for collections of drained CT. Finally, and most obvious, a superficial location is usually necessary for total success. Thus, even now, deep pelvic collections are difficult to drain with ultrasound unless a transvaginal or transrectal approach is used. These routes were not even considered in the late 70s and early 80s, when PAD was just developing.

It is interesting to read some of the early reports on ultrasound-guided abscess drainage, since many of the techniques and procedural issues that today still are held to be important for successful PAD were discussed in the early literature. The major issue and limitation with using ultrasound guidance was, and still is, the ability to select a safe access route. While there are no data that document the frequency of the use of ultrasound guidance in abscess drainage today, it is relatively easy to surmise that over 90% of abscesses are now drained using CT guidance. Interestingly, an early article by MacErlean[10] and associates published in 1981 states that “there appears to be little danger in transfixing organs such as small bowel in order to gain access to an abscess.” Presumably, the authors were alluding to the fact that in their series, the drained abscesses were easily visible. In their series, 3 of 10 abscesses were treated “successfully” with needle aspiration alone. This technique was used when there was bowel between the percutaneous route and the abscess. Clearly, ultrasound in the right hands was successful.

Despite these early reports about abscess drainage using ultrasound guidance, it was the advent of CT that truly revolutionized this procedure. Led in the United States by Gerzof,[11] [12] Haaga,[13] [14] Karlson,[15] Martin,[16] vanSonnenberg,[17] Ferrucci,[18] and Mueller,[18] numerous articles were published in the 1980s establishing the techniques, results, complications, and controversies of PAD. Not long afterward, articles by Bret, Gobien,[19] Dondelinger[20] and Van Waes[21] appeared in the European literature and further certified that PAD was a procedure which would become a revolutionary technique in patient care. Perhaps the most important of these articles was by Gerzof[22] and colleagues, which appeared in the New England Journal of Medicine in September 1981. Then, as now, the worldwide influence and significance of this journal and articles published in it was immense. The article was accompanied by an editorial by world-renowned surgeon Claude E. Welch,[23] who had a 40-year history in abdominal surgery and was influential throughout the world because of his experience and reputation.

Gerzof's[22] article described a 5-year experience with the technique and reported on 71 abscesses in 67 patients. The results described in this article are interesting because they aptly demonstrate the consistency of the technique over a 25-year period. Sixty-one of 71 abscesses (86%) were cured. He had 11 complications, but many were not serious enough that we would even worry about them today: he had a catheter fall out, a local skin infection, and several cases of “septicemia” which are not well described. It is hard to know whether these were “chills” or bacteremia with hypotension. Three of 6 patients who died were not successfully drained. There was one recurrence, and the mean duration of catheter drainage was 20 days.

In this day and age it is difficult to imagine the impact of this article or what Gerzof and the other pioneers had to overcome. It was more complicated than that surgeons did not want anybody else draining abscesses. Questions abounded. Could catheters “that small” actually drain abscesses in a “closed” fashion? Indeed, prior to PAD, surgeons made large incisions and manually washed out abscesses by hand. Could anatomical planes be crossed and not cause superficial or other cross-contamination? The idea that you could place a catheter through the gluteal muscles into the presacral space to drain an abscess was an anathema to a surgeon. Could abscess drainage be successful if the drain was not placed in a dependent position? The idea that an 8- or 10-French catheter could be placed via an anterior approach and actually cure an intraperitoneal abscess in the deep recesses of the abdomen didn't seem possible. Finally, and not to be underestimated, was the fact that few if any radiologists, with the exception of a handful of vascular radiologists who had taken care of GI bleeders, had never taken the responsibility of direct patient interaction, let alone direct patient care.

A look at Dr. Welch's editorial is worthwhile. A couple of quotations from it are given below. He mentions that small-sized catheters might be an issue. “Because of the analogy between abscess and deep wound sepsis, and because of their past experience, surgeons prefer open drainage: this procedure allows evacuation of all locules and detritus, which is often extensive and could block a “small catheter.” He continues, “catheter drainage is generally contraindicated unless the abscess is directly in contact with the abdominal wall.” On pelvic abscess, he states, “pelvic abscesses can readily be evacuated through the rectum by the surgeon, but cannot be drained satisfactorily by a suprapubic (anterior non-dependent approach) catheter.” Finally, he recommends to his fellow surgeons to remain involved with the patient and “be vigilant.” Although he was not completely negative, I believe he was worried about the fact that radiologists really would not take responsibility for their patients. This was not a horrible statement in 1981, because he was concerned that radiologists would simply perform their technique and not follow the patients, a complaint that we face even today. He concludes his editorial with a very strong statement: “ultimately, more can be accomplished through a laparotomy incision than through a catheter.” This is a fascinating window into the history of early abscess drainage. Having worked at the Massachusetts General Hospital (MGH) with Dr. Welch, I can say from personal experience that he was appropriately skeptical, but at the same time supportive. He would let us work on his patients, and he along with his surgical colleagues at the hospital embraced the technique once they saw our enthusiasm and, more important, our successes.

A few early anecdotes might be interesting. Stephen Gerzof assisted me in the first PAD performed at the MGH. One has to realize that, in addition to using what would now be considered extremely primitive imaging, early gray-scale, non-real-time ultrasound, our basic tools for abscess drainage were nonexistent. We had no procedure trays, no well-defined procedural technique; we were really “flying by the seat of our pants.” There were no abscess catheters. Very few of us, if any, knew the difference between trocar and Seldinger techniques. CT was not used in the late 70s at our institution for abscess drainage. The first patient had only an ill-defined collection seen in ultrasound and clinical evidence consistent with an infected collection posterior to the kidney. Radiology nursing, patient monitoring, conscious sedation did not exist. Steve came over to us with his “favorite catheter” which was basically a large-bore 16-French Argyle Ingram trocar catheter (Sherwood Medical Industries, St. Louis, MO). It was designed for bladder decompression and had a 5-mL retention balloon. There was no ultrasound guidance system for needle placement, no CT angled gantry; there was simply Steve's experience and use of “mental triangulation” to angle the catheter in an extraperitoneal approach toward the collection. He guided me, a nervous, inexperienced, but game radiologist, through that first case. I really don't remember whether the patient or I was in more pain, but I do remember we drained over 100 cc of pus and the patient survived. I suspect that experience was not too dissimilar to others' performing this technique for the first time.

It is hard now, 25 years later, to appreciate the skepticism that was present. Not long after that I remember identifying a left subphrenic collection in a postoperative patient by ultrasound. In those days we were not “ready” to offer PAD. Now there would be no question that we would drain it. We did at least identify the collection to the surgeon. A day later he called me up and said he was in the operating room and he “could not find the abscess.” He was wondering if it was actually there. Finally, he did find it. But the skepticism was such that even I doubted myself and whether there was really an abscess when he first called.

Soon after the Gerzof article, several other articles appeared in the literature describing various experiences with abscess drainage from multiple centers. It is not in the scope of this article to describe each specific reference in the literature that paved the way for abscess drainage in a certain anatomical compartment. However, needless to say, many articles about technique, results, and complications were published on abscess drainages in the liver, pancreas, spleen, retroperitoneum, deep pelvic cavity, chest, and mediastinum.

Abscess drainage has progressed to become, perhaps, the most beneficial interventional procedure of all of interventional radiology in terms of its low complications, high percentage of successful results, and patient safety issues. As such, it may be worthwhile to look at several different aspects of abscess drainage in a historical perspective to realize how and why abscess drainage has become such a fundamental procedure in everyday clinical medicine.

REFERENCES

• 1 McFadzen A JS, Chang K PS, Wong C C. Solitary pyogenic abscess of the liver treated by closed aspiration and antibiotics. A report of 14 consecutive cases with recovery.  Br J Surg . 1954;  l41 141-152
• 2 Ogden W W, Hunter P R, Rives J D. Liver abscess.  Postgrad Med . 1961;  30 11-19
• 3 Holm H, Pedersen J F, Kristensen J K, Rasmussen S N, Hancke S, Jansen F. Ultrasonically guided percutaneous puncture.  Radiol Clin North Am . 1975;  13 493-503
• 4 Smith E H, Bartrum R J. Ultrasonically guided percutaneous aspiration of abscesses.  AJR Am J Roentgenol . 1974;  122 308-312
• 5 Doust D B, Quiroz F, Stewart J M. Ultrasonic distinction of abscesses from other intra-abdominal fluid collections.  Radiology . 1977;  125 213-218
• 6 Gronvall J, Gronvall S, Hegedus V. Ultrasound-guided drainage of fluid-containing masses using angiographic catheterization techniques.  AJR Am J Roentgenol . 1977;  129 997-1002
• 7 Tetze E M, Reeves C D, Longerbeam J K. Treatment of liver abscesses.  Am J Surg . 1973;  126 263-270
• 8 Goldberg B B, Pollack H H. Ultrasonic aspiration transducer.  Radiology . 1972;  102 187-189
• 9 Goldman S M, Minkin S D. Renal carbuncle: the use of ultrasound in its diagnosis and treatment.  J Urol . 1977;  118 525-528
• 10 MacErlean D P, Owen A P, Hourihane J B. Ultrasound-guided percutaneous abdominal abscess drainage.  Br J Radiol . 1981;  54 394-397
• 11 Gerzof S G, Robbins A H, Birkett D H, Johnson W C, Pugatch R D, Vincent M E. Percutaneous catheter drainage of abdominal abscesses guided by ultrasound and computed tomography.  AJR Am J Roentgenol . 1979;  133 1-8
• 12 Gerzof S G, Spira R, Robbins A H. Percutaneous abscess drainage.  Semin Interv Radiol . 1981;  16 62-71
• 13 Haaga J R, Alfidi R J, Havrilla T R. et al . CT detection and aspiration of abdominal abscesses.  AJR Am J Roentgenol . 1977;  128 465-474
• 14 Haaga J R, Weinsten A J. CT-guided percutaneous aspiration and drainage of abscesses.  AJR Am J Roentgenol . 1980;  135 1187-1194
• 15 Karlson K B, Martin E C, Franklin E J. et al . Nonsurgical drainage of intra-abdominal and mediastinal abscesses: report of twelve cases.  Cardiovasc Intervent Radiol 198;4:170.
• 16 Martin E C, Frankuchen E I, Neff R A. Percutaneous drainage of abscess: a report of 100 patients.  Clin Radiol . 1983;  28 97
• 17 vanSonnenberg E, Wing V W, Casola G. et al . Temporizing effect of percutaneous drainage of complicated abscesses in critically ill patients.  AJR Am J Roentgenol . 1984;  142 821
• 18 Mueller P R, vanSonnenberg E, Ferrucci J T. Percutaneous drainage of 25 abdominal abscesses and fluid collections: current procedural concepts.  Radiology . 1984;  151 343
• 19 Gobien R P, Young J WR, Curry N S. et al . Computed tomographic guidance of percutaneous needle aspiration and drainage of abdominal abscess.  J Comput Assist Tomogr . 1982;  6 127-133
• 20 Dondelinger R, Meoli S, Kurdziel J C. Aspiration et drainage percutanes des collections throaciques et abdominales sous controle tomodensitometrique.  J Radiol . 1982;  63 75-79
• 21 Van Waes P G M F, Feldberg M AM, Mali W P. et al . Management of loculated abscesses that are difficult to drain: a new approach.  Radiology . 1983;  147 57-63
• 22 Gerzof S G, Robbins A H, Birkett D H, Johnson W C, Pugatch R D, Vincent M E. Percutaneous catheter drainage of abdominal abscesses guided by ultrasound and computed tomography.  AJR Am J Roentgenol . 1979;  133 1-8
• 23 Welch C E. Catheter drainage of abdominal abscesses.  N Engl J Med . 1981;  305 694-695