Volume 9, Number 6, 2008
© Mary Ann Liebert, Inc.
Controversies in Appendicitis*
Derek W. Meeks and Lillian S. Kao
Background: Appendectomy for appendicitis is a common operation that is performed by most general sur-
geons. Despite the large number of appendectomies, numerous management controversies continue.
Methods: Review of the pertinent English language literature.
Results: Controversy persists regarding the use of imaging studies for diagnosis, laparoscopic vs. open surgi-
cal methods, nonoperative management of uncomplicated appendicitis, delayed management of the patient
with phlegmon from severe appendicitis, the necessity for interval appendectomy, and the merits of deferral
of off-hour appendectomy until the following morning. This review identifies the published reports that sup-
port the pros and cons of each controversy.
Conclusions: Class I data generally are lacking to support the resolution of these controversies. Additional well-
controlled clinical trials are necessary to obtain definitive answers.
formed annually in the United States , controversies con-
tinue regarding its optimal diagnosis and management, sev-
eral of which are discussed with a review of the literature.
ESPITE THE FACT that appendicitis is a common disease,
with approximately 250,000 appendectomies per-
Clinical scenario: A 16 year-old previously healthy male patient
presents at 12:00 AM with ? 24 h of abdominal pain, now local-
ized to the right lower quadrant, associated with fever, nausea, and
vomiting. Physical examination reveals tenderness at McBurney
point. The white blood cell (WBC) count is 15,000 /mm3. Should
routine imaging be performed for suspected uncomplicated acute
Controversy exists regarding the optimal diagnostic strat-
egy for uncomplicated acute appendicitis: (1) Clinical history
and physical examination alone; (2) selective imaging in pa-
tients with intermediate clinical suspicion; and (3) routine
imaging of all patients. The main proposed benefit of rou-
tine imaging is a decrease in the misdiagnosis of appendici-
tis, resulting in fewer negative appendectomies. Routine
imaging may theoretically either increase or decrease perfo-
rations attributable to delay in diagnosis, depending on the
effect on the time to the operating room. Furthermore, imag-
ing is not without complications such as allergic reactions,
contrast-induced nephrotoxicity, and radiation exposure as-
sociated with computed tomography (CT).
Multiple retrospective studies (n ? 300 patients) suggest
that use of CT in particular may decrease the negative ap-
pendectomy rate without an increase in the perforation rate
[2-4]. However, a population-based analysis of 63,707 non-
incidental appendectomies by Flum et al. demonstrated no
improvement in these rates with preoperative imaging .
Despite a significant increase in the percentage of patients
who received either CT or ultrasound scans, the authors
identified no change in the rate of ruptured appendices or
misdiagnosed appendicitis over a 10-year period from 1987
to 1997 [5, 6]. The authors speculated that the lack of bene-
fit from preoperative imaging may have been attributable to
the rate of false-positive tests (21.7% for CT and 20.8% for
ultrasonography), which was higher in this analysis than in
single-institution studies . Furthermore, the authors noted
that the advent of laparoscopy did not reduce the rate of neg-
ative appendectomy. In fact, laparoscopy was associated
with a higher rate of misdiagnosis than open surgery among
women of reproductive age .
Observational cohort studies and analyses of administra-
tive databases are subject to multiple biases, namely con-
founding and selection bias. However, there is a paucity of
randomized trials for diagnosing acute appendicitis that
evaluate outcome according to imaging strategy. Hong et al.
randomized 182 adults with intermediate probabilities of
having appendicitis, defined as an Alvarado score of 2 to 8
 (Table 1), to either clinical assessment or mandatory CT
scan . The diagnostic accuracy was similar for the strate-
gies—90% for clinical assessment and 92% for CT. Although
the length of time from presentation to the operating room
Department of Surgery, The University of Texas Medical School at Houston, Houston, Texas.
*Presented at the 27thannual meeting of the Surgical Infection Society, Toronto, Ontario, Canada, April 21, 2007.
was significantly longer in the CT group, there was no dif-
ference in overall length of stay (LOS), hospital charges, or
perforation rates in the two groups. The trial enrolled more
patients than the predetermined sample size, but the power
was determined using predicted accuracies of 95% for CT
and 80% for clinical judgment. Thus, the trial was powered
only to detect a 15% or greater difference in accuracy be-
tween the groups, and may have been underpowered to
detect a difference in perforation rates. Additionally, the au-
thors did not perform an intention-to-treat analysis, exclud-
ing 33 patients from analysis. Nonetheless, the strength of
this trial was the standardization of criteria for obtaining a
preoperative CT scan. Further modification of these criteria
may be necessary given the crossover between the arms.
Lee et al. performed a randomized trial of selective vs.
mandatory imaging in adults with suspected acute appen-
dicitis . The investigators did not standardize the criteria
for selective imaging, but rather left the decision to the dis-
cretion of the examining physicians. The strategy of manda-
tory imaging resulted in more CT scans but decreased rates
of negative appendectomies (3% vs. 14%; 95% confidence in-
terval [CI] for the difference ?3.5%, 26.3%) and perforation
(10% vs. 18%; 95% CI ?8%, 24%) compared with selective
imaging. The observed differences were not statistically sig-
nificant, but the trial was not powered to detect a difference
smaller than 15% in negative appendectomy rates. There was
no difference in the time to operative intervention in the two
groups, although patients who did not receive a CT scan
went to surgery faster. Despite the differences in negative
appendectomy and perforation rates, the authors did not ad-
vocate mandatory imaging for all patients with suspected
appendicitis but concluded that further, larger trials would
be required to determine if there is a subset of patients who
may benefit from routine imaging.
In the absence of definitive data from other randomized
trials, a practical approach to determining the utility of pre-
operative imaging is to base the decision on the pre-test prob-
ability of acute appendicitis and the likelihood ratio (LR) of
the imaging test. The LR describes how much the odds of
having the disease increase if the test is positive (LR?) or
negative (LR?), and is calculated from the sensitivity and
specificity of the test. A positive LR test is defined as the ra-
tio of the sensitivity: (1 ? specificity), and LR? is defined as
(1 ? sensitivity): specificity. Table 2 illustrates the positive
and negative LRs for CT and ultrasonography in adults and
children calculated from a meta-analysis . The post-test
probability of having appendicitis can then either be calcu-
lated or determined using an LR nomogram (Figs. 1 and 2).
Take, for example, a patient with a high pre-test proba-
bility (e.g., 95%), according to the history and physical ex-
amination, of having acute appendicitis (Fig. 1). A positive
CT scan, which has an LR? of 16, increases the post-test
probability to greater than 99%. A negative CT scan, which
has an LR? of 0.06, reduces the post-test probability to 53%.
A post-test probability of greater than 50% may still warrant
operative intervention even with a negative CT scan. Thus,
given that the original clinical suspicion was high, the sur-
geon is unlikely to change his/her management strategy on
the basis of the findings of the scan.
On the other hand, if the pre-test probability of acute ap-
pendicitis is 50%, the results of a CT scan may change the
management strategy (Fig. 2). A positive CT scan would in-
crease the post-test probability to 94%, whereas a negative
scan would decrease the probability to 6%. Thus, preopera-
tive imaging determines the management strategy when the
pre-test probability is in the intermediate range. Although
systems are available for determining the likelihood of ap-
pendicitis, such as the Alvarado score , there are no widely
accepted criteria for accurate determination of the pre-test
probability of acute appendicitis.
In summary, the role of imaging in diagnosing acute un-
complicated appendicitis is yet to be determined. Despite the
high sensitivity and specificity of CT and ultrasonography,
particularly CT, an increase in the use of imaging has not
MEEKS AND KAO 554
TABLE 1.ALVARADO SCORE FOR PREDICTING
Migration of pain
Tenderness in right iliac fossa
Temperature ? 37.3°C
Leukocytosis ? 10,000/mm3
Neutrophils ? 75%
From reference 7.
TABLE 2.SENSITIVITY AND SPECIFICITY OF CT AND ULTRASONOGRAPHY IN ADULTS AND CHILDREN
likelihood ratioImaging Sensitivity (%)Specificity (%)
aLikelihood ratios for positive and negative test results are calculated using the following formulas:
Data from reference 10.
(1 ? Sensitivity)
(1 ? Specificity)
reduced the incidences of misdiagnosis or perforation in
adults. Ultimately, the utility of preoperative imaging in
guiding management strategy depends on the accurate as-
sessment of the pre-test probability of having acute appen-
dicitis. Further study is necessary to characterize those pa-
tients with an intermediate probability who may benefit
most from scanning.
Clinical scenario: A 16 year-old, previously healthy male patient
presents with 36 h of abdominal pain associated with fever, nau-
sea, and vomiting. Examination reveals tenderness at McBurney
point and localized peritoneal signs. The white blood cell count is
25,000/mm3. The decision is made to operate for acute appendici-
tis. Should patients with suspected acute appendicitis undergo lap-
aroscopic or open appendectomy? Does the approach differ for un-
complicated vs. perforated acute appendicitis?
Once the decision has been made to operate on a patient
with suspected appendicitis, the operative approach itself is
controversial. The theoretical advantages of laparoscopic ap-
pendectomy (LA) over open appendectomy (OA) include
smaller incisions with better cosmesis and fewer incisional
complications, faster recovery with shorter hospital LOS and
earlier return to work, and less pain. Suggested disadvan-
tages of LA include longer operating room time and higher
costs. Multiple randomized trials have been performed, but
many contain methodological flaws such as inadequate al-
location concealment, lack of reporting of randomization
method, failure to blind the patient or healthcare provider,
lack of analysis by intention-to-treat, and incomplete follow
up data [11, 12].
A 2004 Cochrane review of 54 studies comparing LA with
OA, of which 45 enrolled adults, demonstrated a significant
reduction in surgical site infections (OR 0.45; 95% CI 0.35,
0.58) with LA. This estimate was derived from more than
5,000 patients; despite differences in laparoscopic technique,
no significant heterogeneity of results was identified. Lapa-
CONTROVERSIES IN APPENDICITIS555
to calculate post-test probability. First, pre-test probability
must be converted to pre-test odds, which is equivalent to
pre-test probability divided by (1 – pre-test probability).
Multiplying the pre-test odds by the likelihood ratio (LR) re-
sults in the post-test odds, which can be converted to post-
test probability. Post-test probability is equivalent to the
post-test odds divided by (1 ? post-test odds). In a patient
with a pre-test probability of 95% of having acute appen-
dicitis, positive CT scan (LR? ? 16) increases post-test prob-
ability to ? 99% (solid line). A negative CT scan (LR? ?
0.06) reduces post-test probability to 53% (broken line).
Pre-test probability and likelihood ratio can be used
5 5 5
(50%) of having acute appendicitis, positive CT scan in-
creases post-test probability to 94% (solid line). Negative CT
scan reduces post-test probability to 6% (broken line). There-
fore, preoperative CT scan may change clinical management.
In patient with intermediate pre-test probability
roscopic appendectomy was associated with a hospital stay
shorter by 1.1 days (95% CI 0.6, 1.5 days), less pain on the
first postoperative day of 9 points of 100 on a visual analog
scale (95% CI 5, 13), and a faster return to normal activity by
6 days (95% CI 4, 8 days) . However, LA was associated
with longer operative time by 12 minutes (95% CI 7, 16 min-
utes) and higher operation and hospital costs (point estimate
not calculated). There was significant heterogeneity in these
analyses, and the authors noted decreasing differences in op-
erative times between LA and OA in recent years. However,
LA was associated with a higher risk of postoperative intra-
abdominal abscess (OR 2.48; 95% CI 1.45, 4.21). The conclu-
sions were that LA is preferred where the technical and sur-
gical expertise is available.
A 2007 meta-analysis of 34 studies comparing LA with OA
included a subgroup analysis of earlier vs. later studies .
The meta-analysis confirmed the findings of fewer surgical
site infections (OR 0.52; 95% CI 0.39, 0.70) and shorter hos-
pital LOS (0.62 days; 95% CI 0.18, 1.05 days) with LA. Com-
parison of pre- and post-2000 studies demonstrated a re-
duction in the difference in operative time from a mean of
20.3 min (95% CI 13.0, 27.4 minutes) to 4.9 min (95% CI 0.24,
9.55 minutes). Costs were not assessed. As reported in the
Cochrane review, intra-abdominal abscesses were more
common with LA; the risk actually increased in studies pub-
lished after 2000. Whether greater experience with laparos-
copy was accompanied by an increase in LA for perforated
appendicitis, which is associated with intra-abdominal ab-
scesses, is unknown. The authors concluded that LA was safe
and effective, but that open surgery still conferred benefits,
in particular, with regard to the likelihood of postoperative
In summary, LA is associated with fewer incisional com-
plications and shorter LOS than OA. With increased experi-
ence, operative times are becoming comparable. Whether in-
hospital costs are tempered by savings as a result of a faster
return to work and fewer incisional complications has not
been evaluated formally by cost-effectiveness analysis. Fur-
ther trials of LA for suspected perforated appendicitis are
necessary, given the potential increase in intra-abdominal
Does Suspected Uncomplicated Acute Appendicitis
Clinical scenario: A 16 year-old previously healthy male patient
presents with three days of abdominal pain associated with fever,
nausea, and vomiting. Examination reveals tenderness at McBur-
ney point. The WBC is 20,000/mm3. A CT scan of the abdomen
and pelvis reveals a right lower quadrant phlegmon and a dilated,
non-filling appendix with an appendicolith. Should early or de-
layed appendectomy be performed for complicated acute appen-
dicitis? Is interval appendectomy necessary after complicated acute
Until recently, appendectomy was accepted as the stan-
dard of care for uncomplicated acute appendicitis. In 1995,
Eriksson and Granstrom published the results of the first ran-
domized trial of antibiotics vs. surgery . In this pilot trial
in 40 patients, one patient treated with antibiotics required
operation during the first admission for peritonitis, whereas
seven patients were re-admitted within one year with re-
current appendicitis that necessitated surgery. Styrud et al.
published the results of a subsequent multi-center random-
ized trial of 252 men between the ages of 18 and 50 with sus-
pected non-perforated acute appendicitis . Fifteen pa-
tients (12%) randomized to the antibiotic arm required
operation within the first 24 h for local peritonitis and fail-
ure to improve, and 16 patients (15%) had recurrences within
a year. The surgical group had a 14% complication rate (17
patients), largely SSIs, whereas the antibiotic group had a 3%
rate (four patients) of complications from surgery after fail-
ing non-operative therapy. There was no difference in the
initial hospital LOS. The authors concluded that acute non-
perforated appendicitis could be treated non-operatively
Critics of this trial cite several methodological flaws .
First, the authors did not define their primary outcome mea-
sure; they compared the 14% complication rate in the surgi-
cal group with the 15% recurrence rate in the antibiotic group
and reported no statistical difference. They did not include
the patients who required operative intervention within the
first 24 h in the antibiotic group (12%) as a failure of non-
operative therapy. Also, the authors did not report the basis
for their sample size calculations; thus, the power of the trial
to detect a difference in complication rates was questioned.
Additionally, the results lack external validity, given that
women and patients older than 50 years of age were ex-
cluded. Furthermore, only 6% of the patients underwent LA,
which is associated with fewer SSIs, limiting generalization
of the trial results to current practice, where laparoscopy is
Farahnak et al. performed a randomized pilot comparison
of a protocol for treating acute appendicitis on the basis of
the Alvarado score with conventional therapy . The pro-
tocol consisted of discharge from the hospital for an Al-
varado score of 0–4 points, antibiotic treatment for 5–7
points, and appendectomy for 8–10 points. Twenty-one pa-
tients were randomized to each group. There was no differ-
ence in non-therapeutic appendectomies or perforation with
the protocol, but there was a significant decrease in hospital
LOS and time to surgery in patients undergo appendectomy.
Most of the patients who underwent appendectomy had an
Alvarado score of 8–10 points in both the case and control
groups. Additionally, the majority of patients who received
antibiotics for a score of 4–7 points may not have had ap-
pendicitis; only one of nine patients underwent an operation
that confirmed the diagnosis. Thus, conclusions regarding
the safety and efficacy of antibiotics for the treatment of ap-
pendicitis are not possible. Furthermore, as a pilot, the trial
was underpowered to detect a difference in negative ap-
pendectomy or perforation rates, and recurrent appendicitis
was not a measured endpoint. However, the trial was de-
signed to be exploratory in order to obtain sample size esti-
mates for a larger trial.
Despite their flaws, these are the only published random-
ized trials evaluating antibiotics without surgery for sus-
pected uncomplicated acute appendicitis. Currently, one ob-
servational trial and two randomized trials in patients over
18 years of age with suspected appendicitis are recruiting
subjects (www.clinicaltrials.gov). Thus, whereas non-opera-
tive therapy can be undertaken safely in patients with sus-
pected non-perforated appendicitis, further study is neces-
sary to clarify which patients are most likely to benefit from
this strategy and to evaluate its cost effectiveness.
MEEKS AND KAO 556
Another area of controversy is when to operate on perfo-
rated appendicitis with or without abscess. Data on the safety
and efficacy of initial non-operative management of compli-
cated appendicitis (perforated, with phlegmon causing a pal-
pable mass, or with abscess) is derived from observational
cohort studies, which are subject to selection bias and con-
founding [17, 18]. There are no published randomized trials
evaluating timing of operation for complicated appendicitis,
although there are three randomized trials in children that
either are recruiting subjects or in the planning phases.
The potential advantages of initial non-operative therapy,
if successful, are a decrease in hospital LOS, complications,
and cost. The disadvantages include potential recurrence
prior to interval appendectomy and a greater incidence of
complications. Associated small bowel obstruction, abscess
necessitating percutaneous drainage or re-operation, or ban-
demia of WBC on admission may be associated with delayed
appendectomy [19, 20]. Selective initial non-operative man-
agement may be an alternative strategy, but predictors of
failure of non-operative therapy must be identified to per-
mit optimal patient selection. Data from the planned ran-
domized trials may elucidate this controversial topic.
Interval appendectomy to prevent recurrence after suc-
cessful initial non-operative management of complicated
appendicitis is controversial as well. Data on the need for in-
terval appendectomy are based on observational cohort
studies and are limited. Proponents of an expectant man-
agement strategy argue that the recurrence rate is low and
that routine interval appendectomy is associated with more
complications and higher cost [21–23]. In a retrospective co-
hort study using discharge data, Kaminski et al. reported a
5% recurrence rate after initial non-operative management
of acute appendicitis, with or without abscess, with a me-
dian four-year follow up . On the other hand, opponents
argue that expectant management may delay the diagnosis
of an occult malignant tumor. In a retrospective study, Lai
et al. estimated the incidence of carcinoma of the cecum as-
sociated with presumptive appendicitis to be 0.85% . Pa-
tients undergoing expectant management therefore may
benefit from colonoscopy or barium enema study to detect
underlying diseases or neoplasm [22, 24].
Can Treatment of Acute Uncomplicated Appendicitis
Be Delayed Until the Following Morning?
On the basis of data suggesting that antibiotic management
of acute appendicitis is safe, as either initial or definitive ther-
apy [14, 16, 21], the routine practice of operating immediately
on acute appendicitis has come into question. The rationale
for immediate operative intervention is to prevent perforation.
Ditillo et al. performed a retrospective review of 1,081 adults
with acute appendicitis  and found that longer time to the
operating room was associated with progressively advanced
pathology, with the spectrum ranging from acute to gan-
grenous to perforated appendicitis or phlegmon to periap-
pendicular abscess. In particular, delay beyond 71 h compared
with less than 12 h was associated with a 13-fold increase in
the odds of advanced pathology. Delay in treatment also was
associated with a higher complication rate and longer hospi-
tal LOS. Similarly, Bickell et al. demonstrated an increase in
the risk of rupture for every 12 h beyond 36 h of symptoms
. Both groups of investigators concluded that delay in op-
eration should be avoided to prevent perforations and a
higher incidence of complications.
Despite the prevalence of acute appendicitis, questions re-
main regarding its optimal diagnosis and management.
Large, multicenter randomized trials are required to answer
Author Disclosure Statement
No competing financial interests exist.
1. Addiss DG, Shaffer N, Fowler BS, Tauxe RV. The epidemi-
ology of appendicitis and appendectomy in the United
States. Am J Epidemiol 1990;132:910–925.
2. Bendeck SE, Nino-Murcia M, Berry GJ, Jeffrey RB Jr. Imag-
ing for suspected appendicitis: Negative appendectomy and
perforation rates. Radiology 2002;225:131–136.
3. Jones K, Pena AA, Dunn EL, et al. Are negative appendec-
tomies still acceptable? Am J Surg 2004;188:748–754.
4. Rao PM, Rhea JT, Rattner DW, et al. Introduction of ap-
pendiceal CT: Impact on negative appendectomy and ap-
pendiceal perforation rates. Ann Surg 1999;229:344–349.
5. Flum DR, Morris A, Koepsell T, Dellinger EP. Has misdi-
agnosis of appendicitis decreased over time? A population-
based analysis. JAMA 2001;286:1748–1753.
6. Flum DR, McClure TD, Morris A, Koepsell T. Misdiagnosis
of appendicitis and the use of diagnostic imaging. J Am Coll
7. Alvarado A. A practical score for the early diagnosis of acute
appendicitis. Ann Emerg Med 1986;15:557–564.
8. Hong JJ, Cohn SM, Ekeh AP, et al. A prospective random-
ized study of clinical assessment versus computed tomog-
raphy for the diagnosis of acute appendicitis. Surg Infect
9. Lee CC, Golub R, Singer AJ, et al. Routine versus selective
abdominal computed tomography scan in the evaluation of
right lower quadrant pain: A randomized controlled trial.
Acad Emerg Med 2007;14:117–122.
10. Doria AS, Moineddin R, Kellenberger CJ, et al. US or CT for
diagnosis of appendicitis in children and adults? A meta-
analysis. Radiology 2006;241:83–94.
11, Sauerland S, Lefering R, Neugebauer EA. Laparoscopic ver-
sus open surgery for suspected appendicitis. Cochrane Data-
base Syst Rev 2004;2:CD001546.
12. Bennett J, Boddy A, Rhodes M. Choice of approach for ap-
pendicectomy: A meta-analysis of open versus laparoscopic
appendicectomy. Surg Laparosc Endosc 2007;17:245–255.
13. Eriksson S, Granstrom L. Randomized controlled trial of ap-
pendicectomy versus antibiotic therapy for acute appen-
dicitis. Br J Surg 1995;82:166–169.
14. Styrud J, Eriksson S, Nilsson I, et al. Appendectomy versus
antibiotic treatment in acute appendicitis: A prospective
multicenter randomized controlled trial. World J Surg 2006;
15. Soreide K, Korner H, Soreide JA. Type II error in a random-
ized controlled trial of appendectomy vs. antibiotic treat-
ment of acute appendicitis. World J Surg 2007;31:871–872.
16. Farahnak M, Talaei-Khoei M, Gorouhi F, et al. The Alvarado
score and antibiotics therapy as a corporate protocol versus
conventional clinical management: A randomized con-
trolled pilot study of approach to acute appendicitis. Am J
Emerg Med 2007;25:850–852.
CONTROVERSIES IN APPENDICITIS 557
17. Brown C, Abrishami M, Muller M, Velmahos G. Appen- Download full-text
diceal abscess: Immediate operation or percutaneous drain-
age? Am Surg 2003;69:829–832.
18. Oliak D, Yamini D, Udani V, et al. Initial nonoperative man-
agement for periappendiceal abscess. Dis Colon Rectum
19. Kogut KA, Blakely ML, Schropp KP, et al. The association
of elevated percent bands on admission with failure and
complications of interval appendectomy. J Pediatr Surg
20. Nadler EP, Reblock KK, Vaughan KG, et al. Predictors of
outcome for children with perforated appendicitis initially
treated with non-operative management. Surg Infect 2004;5:
21. Kaminski A, Liu IL, Applebaum H, et al. Routine interval
appendectomy is not justified after initial nonoperative
treatment of acute appendicitis. Arch Surg 2005;140:897–901.
22. Lai HW, Loong CC, Chiu JH, et al. Interval appendectomy
after conservative treatment of an appendiceal mass. World
J Surg 2006;30:352–357.
23. Lai HW, Loong CC, Wu CW, Lui WY. Watchful waiting ver-
sus interval appendectomy for patients who recovered from
acute appendicitis with tumor formation: A cost-effective-
ness analysis. J Chin Med Assoc 2005;68:431–434.
24. Lai HW, Loong CC, Tai LC, et al. Incidence and odds ratio
of appendicitis as first manifestation of colon cancer: A ret-
rospective analysis of 1,873 patients. J Gastroenterol Hepa-
25. Ditillo MF, Dziura JD, Rabinovici R. Is it safe to delay ap-
pendectomy in adults with acute appendicitis? Ann Surg
26. Bickell NA, Aufses AH Jr, Rojas M, Bodian C. How time af-
fects the risk of rupture in appendicitis. J Am Coll Surg
Address reprint requests to:
Dr. Lillian S. Kao
Department of Surgery
5656 Kelley Street, Suite 30S 62008
Houston, TX 77026
MEEKS AND KAO558