Saline vs Tissue Plasminogen Activator

Irrigations after Drain Placement for

Appendicitis-Associated Abscess:
A Prospective Randomized Trial
Shawn D St Peter, MD, FACS, Sohail R Shah, MD, FACS, Obinna O Adibe, MD, Susan W Sharp, PhD,
Brent Reading, MD, Brent Cully, MD, George W Holcomb III, MD, MBA, FACS, Douglas C Rivard, DO

BACKGROUND: Emerging data suggest instillation of tissue plasminogen activator (tPA) is safe and potentially
efficacious in the treatment of intra-abdominal abscess. To date, prospective comparative data
are lacking in children. Therefore, we conducted a randomized trial comparing abscess
irrigation with tPA and irrigation with saline alone.
STUDY DESIGN: After IRB approval, children with an abscess secondary to perforated appendicitis who had a percu-
taneous drain placed for treatment were randomized to twice-daily instillation of 13 mL 10% tPA or
13 mL normal saline. All patients were treated with once-daily dosing of ceftriaxone and metroni-
dazole throughout their course. The primary end point variable was duration of hospitalization after
drain placement. Using a power of 0.8 and an a of 0.05, a sample size of 62 patients was calculated.
RESULTS: Sixty-two patients were enrolled between January 2009 and February 2013. There were no
differences in demographics, abscess size, abscess number, admission WBC, or duration of
symptoms. Duration of hospitalization after drainage was considerably longer with the use
of tPA. There was no difference in total duration of hospitalization, days of drainage, or
days of antibiotics. However, medication charges were higher with tPA.
CONCLUSIONS: There are no advantages to routine tPA flushes in the treatment of abdominal abscess second-
ary to perforated appendicitis in children. (J Am Coll Surg 2015;221:390e396.
2015 by
the American College of Surgeons)

Abscess formation is the most common complication after
appendectomy for perforated appendicitis in children.1-4
The presence of a well-formed abscess can also be found
in patients presenting with perforated appendicitis.2,5,6 Inser-
tion of a percutaneous drain by interventional radiologists is
a standard treatment for large abscesses.5-9 To facilitate
drainage of these abscesses, there are emerging data
CME questions for this article available at
http://jacscme.facs.org
Disclosure Information: Authors have nothing to disclose. Timothy J Eberlein,
Editor-in-Chief, has nothing to disclose.
Presented at the British Association of Pediatric Surgeons, Edinburgh, Scot-
land, July, 2014.
Received February 25, 2015; Revised March 20, 2015; Accepted March 22,
2015.
From the Departments of Surgery (St Peter, Shah, Adibe, Sharp, Holcomb)
and Radiology (Reading, Cully, Rivard), The Center for Prospective Clin-
ical Trials, The Children’s Mercy Hospital, Kansas City, MO.
Correspondence address: Shawn D St Peter, MD, FACS, Department of
Surgery, The Children’s Mercy Hospital, 2401 Gillham Rd, Kansas City,
MO 64108. email: sspeter@cmh.edu
describing the instillation of fibrinolytic agents into the
abscess cavity to promote faster resolution.10-12 In a prospec-
tive pilot trial in 20 adult patients, there was more rapid and
complete resolution of abscesses using twice-daily instillation
of tPA.11 A recent case series in children with abscesses from a
variety of conditions found tPA flushes after drain placement
to be safe and relatively effective.12 However, there are
currently no comparative data in children. Therefore, we
performed a prospective, randomized trial to evaluate
the effectiveness of fibrinolysis in the treatment of abdominal
abscesses secondary to perforated appendicitis in children.
METHODS
Approval was obtained from the IRB (#09 04 079) before
enrolling patients in this study. Patients were subsequently
enrolled after obtaining permission from their legal guard-
ian. The enrollment process occurred before insertion of
the drain and after diagnosis of a well-formed, apparently
drainable abscess. The permission forms and consent
process were audited by the IRB on a continuing basis.

ª 2015 by the American College of Surgeons http://dx.doi.org/10.1016/j.jamcollsurg.2015.03.043

Published by Elsevier Inc. 390 ISSN 1072-7515/15
Vol. 221, No. 2, August 2015 St Peter et al
The study was registered with clinicaltrials.gov at the
inception of enrollment (NCT00981136).
Participants
The study population consisted of children under age 18
years who were found to have a well-formed abscess
secondary to perforated appendicitis identified by CT.
Families were approached if the radiologist believed that
drainage of the abscess was possible and would benefit
the patient. Patients were excluded if a drain could not
be successfully placed or frank pus was not extracted on
initial aspiration. Patients were also excluded if they were
known to have immune deficiency or another condition
affecting surgical decision making or recovery (eg, hemo-
philia, severe cardiac or respiratory comorbidities) (Fig. 1).
Interventions
Pigtail drains were placed by 1 of 3 interventional radiol-
ogists dictated by the call schedule under sedation using
image-guided percutaneous techniques.
Figure 1. CONSORT (Consolidated Standards of Reporting Trials) flow diagram.
Trial of Irrigation Types in Abscess Cavity 391
Sample size
This was a definitive trial design using duration of hospital-
ization after drain placement as the primary end point vari-
able. Power calculation was based on the mean and SD from
our previous prospective trial on the treatment of an
appendiceal abscess for patients who were managed with a
percutaneous drain.5 In this previous trial, patients present-
ing with a well-formed abscess due to perforated appendicitis
who were treated with initial nonoperative management and
subsequent interval appendectomy, demonstrated a mean
length of hospitalization of 3.6 days after drainage. To offset
the expense of the additional medication and produce net
benefit, the tPA protocol should be able to reduce the
post-drainage hospitalization by at least 1 day. Therefore,
using an estimated SD of 2 days of hospitalization, a power
of 0.8, and significance established at a ¼ 0.05, sixty-two
patients were calculated to be needed for this trial.
Secondary outcomes included total duration of hospi-
talization, days of drainage, amount of drainage, days of
antibiotics, and recurrence of abscess.
392 St Peter et al Trial of Irrigation Types in Abscess Cavity
Assignment
A computer-generated individual unit of randomization
was used in a nonstratified sequence in blocks of 4. After
consent for study enrollment was obtained, the randomi-
zation sequence was accessed in a separate location to
identify the next allotment. The attending surgeons and
interventional radiologists did not obtain consent and
were blind to the allotment throughout the enrollment
process. All data were analyzed on an intention-to-treat
basis, and patients remained in their assigned group.
The rounding teams could not be completely blinded to
randomization, as both the surgical and interventional
radiology teams participated in the drain flushes during
the course of hospitalization.
Protocol
After drain placement, the saline group was managed by
flushing the drain with 13 mL normal saline twice a
day. Our previous clinical protocol was 10 mL saline
twice a day. In preparation for this study, a standard
10F pigtail drain was filled with saline to determine the
volume of the drain, and saline ran out of the holes in
the end of the tube after 3 mL was injected. This means
that 3 mL remains in the drain, as we did not deliver
another flush after instillation. Therefore, a volume of
13 mL assured that 10 mL solution would reach the
abscess. The tPA group underwent the same drain
placement, but the drain was flushed with a solution
containing 1.3 mg tPA in 13 mL normal saline to main-
tain the 10% concentration our institution has used pre-
viously for empyema and abscesses. The drain in both
groups was flushed with the solution at the time of place-
ment, then twice a day when the drain was in place. If the
abscess cavity would not tolerate a volume of 13 mL, then
the amount that could be introduced with gentle manual
syringe pressure was instilled.
Once the drainage was down to 20 mL during the
preceding 24 hours, the drain was removed. If the patient
was clinically well and otherwise met discharge criteria,
but still had drainage >20 mL/d, he or she was discharged
with the drain in place. Discharge criteria, which were the
same for all patients and identical between groups,
included being able to tolerate a regular diet with pain
controlled adequately with oral medications. However,
the earliest the patient could be discharged with the drain
in place was the third day, and at least 6 irrigations needed
to have been performed.
All patients were treated with daily dosing of ceftriax-
one (50 mg/kg) and metronidazole (30 mg/kg), and
completed a 2-week antibiotic course at home through
a peripherally inserted central catheter.
J Am Coll Surg
Data collection
A single individual who had no role in the clinical care,
collected all data prospectively. Abscess size was estimated
by analyzing CT scans for the largest anteroposterior and
lateral size to give the greatest two-dimensional area on an
axial image. Given that the complex shapes and extent of
the abscess cavities made more precise measurements
unfeasible, this two-dimensional measurement served as
a surrogate for size.
Statistical analysis
Continuous variables were compared using an indepen-
dent sample, 2-tailed Student’s t-test. Discrete variables
were analyzed with 2-tailed Fisher’s exact test and
chi-square with Yates correction, where appropriate.
Significance was defined as p  0.05.
RESULTS
Between June 2009 and February 2013, sixty-two patients
were randomized to either saline or tPA flushes after drain
placement for abdominal abscesses secondary to perfo-
rated appendicitis, regardless of whether the abscess
presented before or after appendectomy. There were
1,689 patients with appendicitis during this time frame,
of which 66 patients presented with perforated appendi-
citis and an abscess and were initially managed without
an operation, and abscess after laparoscopic appendec-
tomy for their perforated appendicitis developed in 97
patients. Of these patients, 68 patients underwent percu-
taneous drainage of their abscess. Consent was obtained in
67 patients. However, 5 patients failed successful drain
placement after randomization, so these patients were
excluded. There were no dropouts after enrollment.
Patient characteristics
There were no differences in age, weight, BMI, sex distri-
bution, abscess size, abscess number, admission tempera-
ture, admission leukocyte count, or duration of symptoms
(Table 1). An abscess developed postoperatively after lapa-
roscopic appendectomy for perforated appendicitis in 15
of the saline patients and 7 of the tPA patients (p ¼ 0.07).
Outcomes
There was a significantly longer duration of hospitaliza-
tion after drainage in the tPA group (Table 2). There
was no difference in total duration of hospitalization or
days of drainage. The volume of drainage trended toward
significance, with more volume drained in the tPA group
(p ¼ 0.06).
There was no difference in the total days of IV antibi-
otics or total number of health care visits (Table 3).
Vol. 221, No. 2, August 2015 St Peter et al Trial of Irrigation Types in Abscess Cavity 393

Table 1. Patient Variables on Presentation

Variables Saline (n ¼ 32) Tissue plasminogen activator (n ¼ 30) p Value
Age, y 10.6  3.6 8.7  4.1 0.06
Weight, kg 38.7  14.2 35.1  20.3 0.41
BMI, percentile 48.2  30.4 51.5  36.1 0.70
Sex, male, % 59 57 0.99
WBC, 1,000/mm3 17.5  5.7 18.3  6.1 0.63
Admit temperature,  C 37.7  0.8 37.8  1.0 0.66
No. of abscesses 2.6  1.7 1.6  1.0 0.14
Abscess area, cm2 51.6  39 39.1  24.9 0.17
Data are presented as mean  SD, except where indicated.

Medication charges were clearly greater with the use of
tPA, which was more expensive by a mean of >$2,000
per patient (p ¼ 0.002). There were 2 recurrent abscesses
in the saline group and 6 in the tPA group, which did not
reach significance, and the odds ratio for recurrence with
saline was 0.27 (95% CI, 0.03-1.68). There were higher
mean maximum temperatures in the tPA group on post-
operative days 1 and 2, after which they were similar
(Fig. 2).
One patient could not tolerate tPA secondary to pain
with flushes and was switched to saline on the second
day of treatment, and another who had increased blood
in the drain was transitioned to saline on day 4. However,
in this patient, the drain was removed after only one saline
flush, and the patient was discharged home. Both of these
patients were analyzed in the tPA group. There were no
patients in either group that could not tolerate the full
volume of irrigation.
Due to the uneven distribution of pre- and postopera-
tive abscesses, with more postoperative abscesses in the
saline group, these groups were analyzed separately. Dura-
tion of hospitalization after drainage in patients present-
ing with perforated appendicitis and a well-formed
abscess was 3.4  1.1 days in the saline group and 5.6
 3.4 days in the tPA group (p ¼ 0.01). In patients
with an abscess that developed postoperatively, the results
were 3.2  1.6 postoperative days in the saline group and
3.9  1.1 postoperative days in the tPA group (p ¼ 0.24).
There were 3 patients who were discharged with the
drain in place in the tPA group, 2 on day 3 and 1 on
day 5. One patient was discharged with the drain in the
saline group on day 6.
Because the primary end point was a hospital param-
eter, there were no dropouts affecting the results of the
trial. In follow-up, all patients remained in the system
until complete resolution of the abscess and discontinua-
tion of antibiotics.
DISCUSSION
This study did not demonstrate a clear benefit to fibrino-
lysis in the routine management of abdominal abscesses
secondary to appendicitis using a moderate volume of
flush. In fact, the time to discharge after drain placement
was considerably increased in the tPA group when the
management and discharge criteria were the same between
groups. Given there was a trend toward an increased
volume of drainage with tPA, it might appear that tPA
decreased the viscosity of the drainage, causing more
drainage so the drain stayed longer, which resulted in a
longer duration of hospitalization after drain placement,
and a negative result in the primary end point. Therefore,
it is possible that the increased drainage could still be clin-
ically beneficial by facilitating more complete clearance of
the abscess. However, this was not borne out in the end
results, as we found no differences in total health care
visits, days of antibiotics, or recurrent abscesses between
groups. Data that our group and others have generated
suggest that drainage might not be necessary for all
abdominal abscesses, and appears not to have benefit in
small- to moderate-sized abscesses (up to 20 cm2).13,14
Therefore, it appears that whatever benefit the drain pro-
duces is not likely secondary to complete clearance of the
abscess because some patients recover just as well with no

Table 2. Outcomes of Abscess Drainage

Outcomes Saline (n ¼ 32) Tissue plasminogen activator (n ¼ 30) p Value
Post-drainage days of hospitalization 3.3  1.3 4.5  1.6 0.002
Total initial days of hospitalization 6.4  4.0 7.1  3.8 0.49
Days of drainage 3.6  3.2 4.7  2.5 0.13
Total volume of drainage, mL 128  160 204  166 0.06
Data are presented as mean  SD.
394 St Peter et al Trial of Irrigation Types in Abscess Cavity
Table 3. Outcomes of Complete Medical Course
Medical course Saline (n ¼ 32)
Total health care visits, n 5.2  2.3
Days of IV antibiotics 15.6  4.0
Recurrent abscess, n (%) 2 (6)
Medication charges, $ 4,100  2,600
Data are presented as mean  SD, except where indicated.
drainage at all. As we continue to work on understanding
the ideal management for abdominal abscesses, these data
show that once a drain is placed, there is not an obvious
advantage to the routine use of fibrinolysis, and fibrino-
lytic therapy is more costly.
These results conflict with several articles in the adult
literature. A randomized trial in 20 adult patients with
abdominal abscesses from any source compared twice-
daily infusion of approximately half of abscess volume
of tPA vs saline, with the end point being the degree of
resolution on follow-up CT based on estimated abscess
volumes.11 The study favored tPA with more complete
resolution on imaging and shorter length of stay, although
recurrence was the same, with one in each group. The
heterogeneous causes of the abscesses, length of time the
abscesses were present, and complex medical conditions
in these patients, make it difficult to translate these data
to the pediatric population. Another trial in 100 adults
compared 3-times-daily infusion of 10 mL urokinase
and saline to treat abscesses from any source.15 There
was a substantial reduction in days of drainage and length
of hospitalization with the patients receiving urokinase
flushes. The complexity of the abscesses and the patients
is apparent, as the length of drainage was 8 and 16 days
for urokinase and saline, respectively. These findings are
also difficult to compare with our study, with a common
abscess source in healthy children. In our study, the mean
days of drainage were 4.6 vs 3.5 days for tPA and saline,
respectively. The fact that the duration of drainage with
Figure 2. Mean maximum daily temperatures between groups for the first 5 days. tPA, tissue
plasminogen activator.
J Am Coll Surg
Tissue plasminogen activator (n ¼ 30) p Value
5.9  2.3 0.24
16.8  5.0 0.3
6 (20) 0.22
6,500  3,100 0.002
saline irrigations was identical to our previous results
with saline irrigations validates that this is a realistic rep-
resentation of our current capacity to manage patients.5
There were more recurrences in the tPA group, which,
although not significant, was a concerning trend. Given
the large shift in percentages created by single events in
a sample of this size, less value can be placed on there be-
ing more recurrences in the tPA group. However, these
data strongly suggest that recurrence rates are not better
with tPA in this model. The adult studies have found
no differences in recurrences.11,15 Likewise, the fact that
these results were generated despite the fact that the
number of abscesses and size of abscesses trended toward
being greater in the saline group, also decreases the likeli-
hood that the findings of this study are spurious.
In our current study, the comparison is between a rela-
tively costly medication with powerful physiologic effects
against saline. Therefore, a trial showing no difference is
more significant because an advantage with a clinically
meaningful effect size should be found to recommend
the routine use of the more expensive and higher-risk
treatment. Instead of designing the trial with larger
numbers to identify a smaller difference in outcomes, it
is intuitive that the difference should be substantial to
offset the expense of tPA. The trial was designed to detect
a 1-day difference in hospitalization after drainage, and
this was precisely the difference that was found, but in
favor of saline. In addition, there were higher mean
maximum temperatures on hospital days 1 and 2 in the
Vol. 221, No. 2, August 2015 St Peter et al
tPA group, and the number of recurrent abscesses in the
tPA group is worrisome. Although the small numbers
prevented this difference from being significant, the direc-
tion of effect is negative for tPA. It is unlikely that patient
differences would account for this finding, given that all
the measured parameters of illness were comparable at
presentation (Table 1).
The distribution of pre- and postoperative abscesses
might be important because there were more postopera-
tive abscesses in the saline group. Theoretically, this
should have provided an advantage to the saline group
because there were more patients who had been treated
with IV antibiotics for a week or more before diagnosis
of the abscess that triggered enrollment. The fact that
the tPA group had more patients with a presenting abscess
who had received no previous treatment could have been a
disadvantage. However, subset analysis comparing out-
comes for abscesses diagnosed before or after appendec-
tomy revealed the opposite, as saline outperformed tPA
far more in those with a preoperative abscess, of which
there were fewer in the saline group. Therefore, the pre-
and postoperative abscess distribution among groups
seems to exaggerate the results instead of tempering them.
This study was inclusive for all drainable abscesses
resulting from perforated appendicitis, which invariably
results in a spectrum for the length of time the abscess
had been maturing and the amount of solid or septated
components. Theoretically, the advantage of tPA would
be to break down septations. The ability to facilitate
drainage by decreasing viscosity of thick, septated, and
solid purulent components has been the standard ratio-
nale for the use of tPA.12,15-19 Therefore, our results do
not exclude the possibility that tPA could offer therapeu-
tic advantages in more complex, septated abscesses. In
addition, the adult trials used larger volumes or more
frequent irrigations, so it might be possible that more
aggressive irrigations with tPA could yield different re-
sults. In addition, as the adult trials included patients
with an abscess from any source, we make no assumptions
about other abscess models with a different etiology and
natural history, such as Crohn’s disease, surgical site infec-
tions, and others.
CONCLUSIONS
Routine tPA flushes in drains placed into abscesses associ-
ated with appendicitis in children offered no advantages
over saline.
Author Contributions
Study conception and design: St Peter, Rivard
Acquisition of data: Sharp
Trial of Irrigation Types in Abscess Cavity 395
Analysis and interpretation of data: St Peter, Sharp,
Rivard
Drafting of manuscript: St Peter
Critical revision: St Peter, Shah, Adibe, Sharp, Reading,
Cully, Holcomb, Rivard
Acknowledgment: The authors thank Drs Trish Valusek,
Carissa Garey, Carrie Laituri, Frankie Fike, Vince
Mortellaro, Alessandra Gasior, E Marty Knott, David Juang,
Corey Iqbal, and Pablo Aguayo whose efforts made the
completion of this study possible.
REFERENCES
1. St Peter SD, Adibe OO, Iqbal CW, et al. Irrigation versus suc-
tion alone during laparoscopic appendectomy for perforated
appendicitis: a prospective randomized trial. Ann Surg 2012;
256:581e585.
2. Blakely ML, Williams R, Dassinger MS, et al. Early versus in-
terval appendectomy for children with perforated appendicitis.
Arch Surg 2011;146:660e665.
3. Fraser JD, Aguayo P, Leys CM, et al. A complete course of intra-
venous antibiotics versus a combination of intravenous and oral
antibiotics for perforated appendicitis in children: a prospective,
randomized trial. J Pediatr Surg 2010;45:1198e1202.
4. St Peter SD, Tsao K, Spilde TL, et al. Single daily dosing cef-
triaxone and metronidazole vs. standard triple antibiotic
regimen for perforated appendicitis in children: a prospective
randomized trial. J Pediatr Surg 2008;43:981e985.
5. St Peter SD, Aguayo P, Fraser JD, et al. Initial laparoscopic
appendectomy versus initial nonoperative management and
interval appendectomy for perforated appendicitis with
abscess: a prospective, randomized trial. J Pediatr Surg 2010;
45:236e240.
6. Keckler SJ, Tsao K, Sharp SW, et al. Resource utilization and
outcomes from percutaneous drainage and interval appendec-
tomy for perforated appendicitis with abscess. J Pediatr Surg
2008;43:977e980.
7. Chen C, Botello C, Cooper A, et al. Current practice patterns
in the treatment of perforated appendicitis in children. J Am
Coll Surg 2003;196:212e221.
8. Morrow SE, Newman KD. Current management of appendi-
citis. Semin Pediatr Surg 2007;16:34e40.
9. Owen A, Moore O, Marven S, et al. Interval laparoscopic
appendectomy in children. J Laparoendosc Adv Surg Tech
2006;16:308e311.
10. Beland MD, Gervais DA, Levis DA, et al. Complex abdominal
and pelvic abscesses: efficacy of adjunctive tissue type plasmin-
ogen activator for drainage. Radiology 2008;247:567e573.
11. Cheng D, Nagata KT, Yoon HC. Randomized prospective
comparison of alteplase versus saline solution for the percuta-
neous treatment of loculated abdominopelvic abscesses. J Vasc
Interv Radiol 2008;19:906e911.
12. Shawyer AC, Amaral JG, Langer JC. The role of tissue
plasminogen activator in the management of complex intra-
abdominal abscesses in children. J Pediatr Surg 2012;47:
1380e1384.
13. Gasior AC, Knott EM, Ostlie DJ, St Peter SD. To drain or not
to drain: an analysis of abscess drains in the treatment of
appendicitis with abscess. Pediatr Surg Int 2013;29:455e458.
396 St Peter et al Trial of Irrigation Types in Abscess Cavity
14. Piper GH, Derinkuyu B, Koral K, et al. Is it necessary to drain
all postoperative fluid collections after appendectomy for
perforated appendicitis? J Pediatr Surg 2011;46:1126e1130.
15. Loborda A, De Gregorio MA, Miguelena JM, et al. Percuta-
neous treatment of intra-abdominal abscess urokinase versus
saline serum in 100 cases using two surgical scoring systems
in a randomized trial. Eur Radiol 2009;19:1772e1779.
16. McRitchie DI, Cummings D, Rotstein OD. Delayed
administration of tissue plasminogen activator reduces
intra-abdominal abscess formation. Arch Surg 1989;124:
1406e1410.
J Am Coll Surg
17. Diamond IR, Wales PW, Connolly B, et al. Tissue plasmin-
ogen activator for the treatment of intra-abdominal abscesses
in a neonate. J Pediatr Surg 2003;38:1234e1236.
18. Sonnappa S, Cohen G, Owens CM, et al. Comparison of uro-
kinase and video-assisted thoracoscopic surgery for treatment
of childhood empyema. Am J Respir Crit Care Med 2006;
174:221e227.
19. St Peter SD, Tsao K, Spilde TL, et al. Thoracoscopic decorti-
cation vs. tube thoracostomy with fibrinolysis for empyema in
children: a prospective, randomized trial. J Pediatr Surg 2009;
44:106e111.