Percutaneous Abscess Drainage

Updated: Nov 16, 2023
  • Author: Evan J Samett, MD; Chief Editor: Kyung J Cho, MD, FACR, FSIR  more...
  • Print


One definition of an abscess is an infected fluid collection that is drainable. A common dictum is as follows: "If it will not go through a catheter, it cannot be drained; if it is not infected, it is not an abscess."

Differentiating a phlegmon from an abscess can be difficult. A phlegmon may be defined as a vascularized infection that still has perfusion. Some nonsuppurative lymphadenitis may not demonstrate enhancement on imaging studies (eg, computed tomography [CT], magnetic resonance imaging [MRI], or amplitude Doppler ultrasonography [US]), but other collections that do not show central enhancement suggest the presence of an abscess, hematoma, or necrosis.

A phlegmon is an undrainable infection. A viscous abscess without significant debris may be difficult to aspirate through a needle but should be drainable through catheters of appropriate caliber. Irrigation with saline or fibrinolytic agents may be necessary for successful drainage of an abscess with significant debris, blood, or viscous elements. [1]

Percutaneous abscess drainage (PAD), [2]  once revolutionary, has evolved into a routine procedure, replacing open surgical abscess drainage in all but the most difficult or inaccessible cases. [3, 4] It was originally believed that only patients with simple fluid collections were candidates for PAD; however, researchers subsequently demonstrated that both septated and viscous fluid collections could be successfully treated percutaneously, particularly with the adjunctive use of lytic agents. (As an example, the mucolytic agent acetylcysteine has been used for multiloculated liver abscesses. [5] ) In general, the simpler the abscess, the more likely it is that PAD will be rapidly successful.

An aggressive practical approach with relatively simple devices and techniques may yield a high success rate with few complications.


Technical Considerations

Best practices

Certain basic medicolegal principles apply to all invasive radiology, and thus to PAD as well. [6] Invasive radiology lawsuits are usually related to procedure complications rather than to misdiagnosis. Complaints stem from the following five general causes:

  • Performance of a procedure that was not indicated - It is usually alleged that a less invasive or less risky procedure could have been performed; when appropriate, there should be chart documentation of the indication of the procedure performed (vs alternative approaches) by the radiologist and referring physician
  • Failure to obtain informed consent - The physician must provide the patient with a general understanding of the procedure, its risks and hazards, and the medically acceptable alternatives, and consent for all potential procedures must be obtained in advance before a patient receives sedation-analgesia; for underage or incompetent patients, the responsible consenting person must be known; the physician must recognize when a patient is withdrawing consent during a procedure and comply
  • Failure to perform the procedure in a reasonable manner and deviation from the standard of care - The radiologist should have documentation of adequate training, including a logbook of procedures, successes, and complications, and there should be documentation of adequate training and continuing medical education (CME) for new procedures and technology
  • Failure to promptly recognize and react to a complication - Satisfactory and unsatisfactory results of procedures must be promptly and adequately noted in the chart, and follow-up care should be documented
  • Failure to adequately treat the complication according to an adequate standard of care - The radiologist must be familiar with the recognition and treatment of contrast reactions; basic life support and/or advanced cardiac life support certification should be kept current; there should be documentation of consultations obtained (eg, surgical) in the treatment of complications

Complication prevention

Successful PAD depends on correct initial placement of the drainage device and on the device remaining in place for the duration of treatment. Accordingly, the greatest complication risk for PAD, as for all catheter placement procedures, is catheter dislodgment.

No ideal fixation device or technique is available. Locking sutures on drainage catheters may not reduce the incidence of catheter dislodgment but are indicated for maintaining a catheter in a cavity removed from the access (as in transhepatic cholecystostomy or transrectal pelvic abscess drainage). [7] Torque generated by a patient rolling over in bed overcomes any catheter fixation system. A catheter fixator is merely a catheter position reminder.

The author has had some success using two catheter fixation devices in tandem. However, even this aggressive approach will not prevent all catheter dislodgments. Debilitated, confused, and uncooperative patients are at risk. Patients often do well in a home setting with well-motivated caregivers.

It is important to educate the patient, as well as clinical nursing and physician staff, to monitor the catheter carefully for possible dislodgment. Catheter dislodgment is reflected in a change in the length of catheter visible outside of the patient. When this is noticed, the catheter should be secured to the skin, and radiographs should be obtained for comparison with the baseline study.

Catheter dislodgment may be noticed only when the catheter is leaking (coiled under the bandage) or on the floor. If it is detected early enough (usually < 8 hr), replacement catheters may be placed along the existing tract by using a 5-French dilator, contrast, and a hydrophilic guide wire.

The realities of healthcare reimbursement and reform have led to patients being discharged from the hospital sooner than before, either to home or skilled nursing facilities. These patients are still in the active phase of PAD but are no longer in a septic state and are cleared for discharge and transfer. As a result, the intensive follow-up that is optimal for abscess resolution may not be as easy to achieve as it once was.

The author has encountered anecdotal cases of patients returning from skilled nursing facilities with larger, more extensive abscesses than they had when they were discharged. This can be attributed to fragmentation of care, with different teams taking over at each facility. In the best of situations, communication with the new treatment team or outpatient follow-up with the original treatment team continues until abscess resolution.



In a study (N = 35) of CT-guided percutaneous catheter drainage for acute necrotizing pancreatitis, Mortelé et al found that primary CT-guided percutaneous drainage alone was successful in approximately 50% of patients (17/35). [8]  The effectiveness of this approach in patients with sterile necrosis was not significantly different from that in patients with infected necrosis. Of 11 patients with multisystem organ failure (10 with sterile necrosis, one with infected necrosis), only four were treated successfully with CT-guided drainage alone; five patients died. Of 24 patients without multisystem organ failure, 13 were treated successfully with CT-guided percutaneous drainage alone; one patient died.

According to a study by Weber et al (N = 44), percutaneous transhepatic biliary drainage (PTBD) should be considered the treatment of choice in patients with benign anastomotic stricture after bilioenterostomy, especially after stricturing of a hepatojejunostomy. [9]  In 27 (61.4%) of the 44 patients, the percutaneous transhepatic biliary drain was successfully removed after 19.9 ± 16.1 months. [9] During follow-up (mean, 53.7 ± 28.4 mo after drain removal), there was no evidence of recurrent strictures. Permanent drains were necessary in 10 patients. In seven patients, repeat operation was necessary because of PTBD failure.

Kloek et al (N = 101) compared the outcomes of endoscopic biliary drainage (EBD) and PTBD in patients with resectable hilar cholangiocarcinoma (HCCA), of whom 90 underwent EBD and 11 PTBD. [10] The technical success rates were 81% for EBD and 100% for PTBD. Stent dislocation was similar in the EBD and PTBD groups; infectious complications were significantly more common in the EBD group; and patients in the EBD group underwent more drainage procedures and had a significantly longer drainage period until laparotomy. In 30 patients, EBD was converted to PTBD because of EBD failure.

Yamakado et al (N = 12) evaluated the safety, feasibility, and clinical utility of PTBD under real-time CT guidance for drainage of inaccessible abdominal abscesses. [11, 12] (Abscesses were considered inaccessible because they were surrounded by the liver and other organs.) An 8-French catheter was advanced into the abscess cavity through the liver parenchyma. Drainage catheters were placed with no complications in all patients, and all abscesses were drained, shrinking immediately after catheter placement.

Pugmire et al (N = 25) assessed the success rate of PAD for abscesses related to Crohn disease in pediatric patients, paying particular attention to end points relevant to biologic therapy. [13] Success was classified as either technical (ie, catheter placement within the abscess with reduction in abscess size on posttreatment imaging) or clinical (either no surgery within 1 year of drainage or surgical resection following drainage with no residual abscess at surgery or on preoperative imaging). All cases were classified as technical successes, and 19 were considered clinical successes.

Ye et al (N = 74), in a study comparing the outcomes of posterior and anteroposterior approaches to percutaneous drainage of tubercular psoas abscesses, found that the posterior approach appeared to have the same clinical efficacy as the anteroposterior approach but was associated with a shorter average hospital stay and a lower complication rate. [14]

A study by El-Hussuna et al (N = 335) assessed risk factors for postoperative complications and the optimal time interval for surgical intervention after percutaneous drainage of intra-abdominal abscesses in patients with active Crohn disease. [15]  Postoperative complications were significantly associated with smoking, low serum albumin, and older age. The risk of abscess recurrence was reduced when there was an interval of at least 2 weeks between successful drainage and subsequent surgery.

 In  a systematic review and meta-analysis (10 studies; N = 1287), Lin et al assessed the effectiveness and safety of percutaneous catheter drainage for liver abscess against that of percutaneous needle aspiration. [16]  The former technique was found to yield a higher treatment success rate, better clinical outcomes by 3 days, and less need for intravenous antibiotics by 4 days.