Preoperative CT imaging as an accurate diagnostic modality for negative laparotomy for patients with acute right lower abdominal pain

Preoperative CT imaging as an accurate diagnostic modality for negative laparotomy for patients

with acute right lower abdominal pain

Taher H Elwan,a MD; Mokhtar Abdelrhman Bahbah,a MD;

Ashraf M Abd elkader,a MD; Islam MEl shaz/y,b MD

a) Department of General Surgery, Benha University, Benha, Egypt. b) Department of Radiology, Benha University, Benha, Egypt.

Abstract

Objectives: Evaluation of the diagnostic  yield of preoperative abdominal  CT imaging of patients presenting  by acute right lower abdominal  pain and the probability for reduction of negative appendectomy rates in these patient.

Patients & methods: The present study aimed to include patients presenting with acute right lower abdominal  pain who were admitted as acute abdominal emergency  under observation. All enrolled patients underwent clinical examination and laboratory investigations to provide a probable clinical diagnosis.All patients underwent abdominal ultrasonography (US) and then scanning with multi-detector  row CT. All patients with clinical suspicion for need of surgical exploration underwent laparotomy and excised specimens were examined pathologically. Patients who became stable with negative US and/or CTwere maintained under-observation till 24 hours and were discharged.

Results: One hundred and seventeen patients underwent surgical exploration including 81 patients who had emergency surgery and 36 who had surgery on elective basis during observation period. Pathological examination confirmed positive diagnosis of acute abdomen in 78 patients and 39 patients were pathologically free (Negative laparotomy).Clinical  diagnosis defined 79, abdominal US defined  83 and CT defined  74 patients  as having  positive  acute abdominal condition. Preoperative CT showed a significantly higher test validity characters in comparison to abdominal  US and clinical examination with sensitivity rate of 94.6%,  specificity  rate of

90.7% and accuracy rate for diagnosis of93.2%. Statistical analysis defined preoperative CT

as the best predictor for negative laparotomy.

Conclusion: Preoperative CT for patients with acute right lower abdominal pain reduces the negative laparotomy rate, improves true positive surgical rate and is mandatory especially in suspicious cases. Moreover, preoperative CT could help di.fforential diagnosis of the underlying pathology and so can modify surgical decision.

Key words: Preoperative, CT, acute right lower abdominal pain, negative laparotomy.

Introduction:

Despite being an ancient disease that dates since early history, diagnosis and proposed lines of management for right lower abdominal pain which is mostly due to appendicitis is still a matter of research.The achievement of proper diagnosis irrespective of the cost of methods used could reduce the consumption of resources through reduction of the rate of negative surgery

with its consequences as reduction of days off work and allowing direction of health resources towards the ideal target.l-3

The diagnosis of acute appendicitis  is not always clear clinically since the specificities of the classic clinical symptoms (periumbilical pain migrating to the right lower  quadrant, nausea, and anorexia) range from 37% to 53%. If acute appendicitis  is  diagnosed  as simple

appendicitis, the recovery time is relatively short time without any complications. However, consequences of missing appendicitis are severe, in perforated or gangrenous appendicitis due to the delay of operation, the hospital stay, the cost, and the incidence of early, as well as delayed complications are increased drastically in  terms  of  morbidity and  mortality.4,5

Improvement of radiological evalution and evolution of new inflammatory markers allowed higher incidence of early diagnosis and treatment for acute appendicitis, but the incidence of early detection for complicated and/or malpresented appendicitis is still limited.3,6-8

On the other hand, historically, negative appendectomy rates of20%-25% and as high as 40% in women have been considered acceptable. The magnitude of the problem of negative appendectomy is much higher in special situations as abnormal presentation, abnormal site of pain or site of pain referral and in children. Women showed generally higher negative appendectomy rates due to gynecologic disease, which can confound the diagnosis of appendicitis. Also, pregnancy is an important cause  for high  negative laparotomy  for   acute  abdomen.9-12

The current study aimed to evaluate the diagnostic yield of preoperative abdominal CT imaging of patients presenting by acute right lower abdominal pain and the probability for reduction of negative appendectomy rates in such patient population.

Patients and methods:

The  present  prospective  study   was

conducted at General Surgery Department, Benha University Hospital since May  2010 till May 2012. After approval of the study protocol by the Local Ethical Committee and obtaining written fully informed patients' or nearest relative consent, the study included

117  patients presenting  to  Emergency

Department with acute right lower abdominal quadrant pain and were admitted as acute abdominal  emergency  under  observation.

All enrolled patients were clinically examined for demographic and constitutional data. History of recurrent pain, nausea and/or vomiting, fever, constipation or diarrhea was

undertaken. Clinical data including tenderness, rebound tenderness and special clinical signs were determined. Blood samples were taken for complete blood counting. Probable clinical diagnosis was determined depending on clinical and laboratory data.

All patients underwent abdominal ultrasonography and then scanning with multi­ detector row CT (HiSpeed Advantage or Light­ Speed; GE Medical Systems, Milwaukee, Wis) with 1.25- or 5-mm section thickness. Scans obtained at 1.25-mm collimation were reconstructed to  5-mm section  thickness. Intravenous contrast material (iohexol, Omnipaque 350; Nycomed Amersham, Princeton, NJ) 150 ml was injected at a rate of 3-5 ml/sec.

All patients had clinical suspicion for need of surgical exploration were managed emergently after adjustment of general condition especially if there is fever, nausea, vomiting or dehydration. Otherwise, patients who became stable with negative US and/or CT were maintained under-observation till24 hours and were discharged.Excised specimens were examined  pathologically as a gold standard for comparison of diagnostic accuracy.

Statistical analysis:

Obtained data were presented as mean±SD, ranges, numbers and ratios. Results were analyzed using paired Z-test and Chi-square test. Sensitivity & specificity of diagnostic modalities as predictors for negative laparotomy were evaluated using the receiver operating characteristic (ROC) curve analysis judged by the area under the curve (AUC) and Regression analysis (Stepwise method). Statistical analysis was conducted using the SPSS (Version 15,

2006) for Windows statistical package. P value

<0.05 was considered statistically significant.

Results:

The study included 147 patients presented

to ED with a picture suggestive of acute abdomen. One hundred and seventeen patients (79.6%) underwent surgical exploration including 81 patients who had emergency surgery and 36 who had surgery on elective basis during the follow-up period. There were

52 males and 65 females with mean age of

34.4±7.5; range: 19-45 years.  Patients' demographic, constitutional and clinical data at time of enrollment are shown in Table(l).

Clinical diagnosis depending on presenting

symptoms, objective findings and laboratory data, defined 79 patients (67.5%) as having acute abdominal  condition mostly acute appendicitis, while the remaining 38 patients (32.5%) were kept  under  observation. Abdominal US defined 83 patients (70.9%) as having acute abdominal condition, while the remaining 34 patients (29.1%) were ultrasonographically free.  Abdominal  CT defined 74 patients (63.2%) as having acute abdominal, while the remaining 43 patients were free. All of the 117 patients underwent surgical exploration and operative findings and pathological examination confirmed positive diagnosis  of acute abdomen in 78 patients (66.7%) and 39 patients (33.3%) were pathologically free (Negative Laparotomy), Table(2).

Considering pathological diagnosis as the gold standard, preoperative CT showed a significantly higher test validity characters in comparison to abdominal US and clinical examination with sensitivity rate of 94.6%,

Table (1): Studied patients data.

specificity rate of 90.7% and accuracy rate for diagnosis  of 93.2% , Table(3), Figure(l).

Using  ROC  curve  for  evaluating the predictability of clinical diagnosis, ultrasonographic diagnosis and CT imaging as predictors for negative diagnosis showed that both US and CT imaging are significant sensitive predictors for negative laparotomy, Figure(2). Moreover, regression analysis defmed preoperative CT as the best predictor for  negative laparotomy, Tables(4&5).

Concerning differential diagnosis of cause of acute abdomen; CT imaging defined 20 cases of gynecological  emergencies  which were  misdiagnosed depending on US examination, Figures(3-5). Moreover, CT imaging defined 3 cases of acute diverticulitis without abscess formation in two, but the 3rd was diverticular abscess Figures(6&7). One case of cancer caecum, Figure(8)  forming mass mimicking appendicular  mass on US examination was diagnosed on CT imaging. Three cases had impacted stone lower end right ureter, Figure(9). Thirteen cases had complicated appendicitis, Figure(lO), while

38 cases  had uncomplicated appendicitis,

Figures(ll-13).

Data are presented as mean±SD & numbers; ranges & percentages are in parenthesis.

Table  (2): Patients' distribution according to clinical  and radiological diagnosis of acute appendicitis compared versus pathological diagnosis.

Table (3): Test validity characters of clinical and radiological diagnosis of acute appendicitis as the cause of acute abdomen.

PPV: Positive predictive value                NPV: Negative predictive value

Table (4):ROC curve analysis of diagnostic yield of clinical and radiological data for prediction of negative laparotomy.

AUC: area under curve                                                          Std. Error: standard error

Table (5): Regression analysis  of diagnostic procedures as the best predictor for prediction of negative laparotomy.

ft: standardized coefficient                 t: paired t-test                 Sig.: significance

Sensitivity          Specificity               PPV       NPV      Accuracy Figure (1): Test vilidity characters of studied diagnostic   g  nlca procedures for diagnosis of pathological acute abdomen. •cT

LO

0 8

l'

-:: 0.0

i

t:

I,

Q_A

0.2

-Cilnlcal

-us

-C T

0.0

0.0                     0.2                      04 4                        0 6

1 - SpoclficH:y

0.8                       1. 0

Figure (2): ROC curve analysis for the sensitivity of clinical diagnosis, US scanning and CT

imaging for prediction of negative laparotomy.

Figure (3): CT image showing complicated right ovarian cyst, operative exploration revealed hemon-hage in a right side ovarian cyst.

-&iVHI!Aif!)iJ,JfM#NJi:f'

Figure (4): CT image showing right tubo­ ovarian abscess, early stage tubo-ovarian abscess which was drained.

Figure (5): CT image showing complicated ovarian cyst, operative exploration revealed hemorrhage andtorsionof right side ovarian cyst.

Figure(6):CTimageshowing pelvicabscess.

Figure(7a):ShowsCT scan of a woman with sigmoid diverticulitis and colosalpingeal fistula. Axial contrast-enhanced CT scan of pelvis shows normal size of right adnexum, enlarged left adnexum containing fluid and foci of gas.

Figure (8): CT image shows asymmetrical mural thickening of the cecum in a case of cecal cancer, resembling an inflammatory mass.

Figure(9): CTimageshowingimpacted stone lower end right ureter.

Figure (10): CT image showing an area of ill-defined and variable enhancement with pockets of extraluminal gas is present due to an appendiceal abscess. Note that there is thickening  of   the   peri-cecal  fascia.

Figure (12): CT showing enlarged appendix with thickened wall and dilated lumen and appendiceal  wall enhancement,· a picture indicting non-complicated appendicitis.

Discussion:

Reduction of negative appendectomy rate was evident on application of preoperative CT where the frequency of false positive diagnosis of appendicitis was reduced from about 19% depending  on clinical  judgment and/or

abdominal US to 3.4% and the frequency of true negative diagnosis of appendicitis was raised from about 14% and 1911/o depending on clinical judgment  or abdominal  US, respectively to 33.3%, thus reliance on preoperative CT could spare surgery in about one-third of examined patients with accuracy of diagnosis of93.2%.

Considering pathological diagnosis as the gold standard, preoperative CT showed a significantly higher test validity characters in comparison to abdominal US and clinical

-&iVHI!Aifj)iJ,JfiWi»ii:l'

Figure (11): CT showing enlarged appendix with thickened wall and dilated lumen; a picture indicting non-complicated appendicitis.

Figure (13): CT showing enlarged appendix with thickened wall and dilated lumen and periappendicealfatstnding,·a pictureindicting non-complicated appendicitis.

examination with sensitivity rate of 94.6%, specificity rate of90.7% and accuracy rate for diagnosis of 93.2%. Moreover, regression analysis defined preoperative CT as the best predictor for pathological acute abdominal pain, despite the significantly wider area under curve for abdominal US as detected by ROC curve analysis.

These data and reported figures go in hand with Krajewski et ai.t3 who reported that the negative appendectomy rate was 8.7% when using CT compared with 16.7% when using

clinical evaluation alone with significantly lower negative appendectomy rate during the CT era compared with the pre-CT era and concluded that routine CT in all patients presenting with suspected appendicitis could reduce the rate of unnecessary surgery without

Rosen et al.14 documented that among adult patients presenting with clinical signs of acute appendicitis, the sensitivity and specificity of CT are greater than those of ultrasound, with improved performance when CT is performed with  intravenous contrast. Poletti et  ai.lS evaluated an algorithm integrating ultrasound and low-dose unenhanced CT with oral contrast medium in the assessment of acute appendicitis, to reduce  the need of conventional CT and documented that  the  proposed algorithm achieved high sensitivity and specificity  for detection of acute appendicitis, while reducing the need for standard CT and thus limiting exposition to radiation and  to intravenous contrast media.

Petroianu16 documented that the advances

in  imaginology tend  to  diminish the  false positive or negative diagnosis. In experienced hands, ultrasound may have a sensitivity of

90% and specificity higher than 90%. Helical

CT has reported a sensitivity  that may reach

95% and specificity higher than 95%. Despite all medical advances,  the diagnosis of acute appendicitis continues to be a medical challenge.

In support of the accuracy  of abdominal CT for improving outcome of cases of acute abdominal pain; Talanow et aJ.17 described a case with left sided flank pain, workup for nephrolithiasis was negative for renal stones or hydronephrosis, after discharge, the patient presented one week later in the ED with right lower quadrant pain and contrast enhanced CT of  the   abdomen revealed perforated appendicitis. Abo et ai.lB tried to determine the  relationship between  body  mass  index (BMI) and accuracy of US and CT scan for suspected appendicitis and reported that the sensitivity and specificity of CT for appendicitis are excellent  regardless of BMI, while there is a trend  of decreasing sensitivity with increasing BMI when using US. Tawk et al.19 described  a case presenting with a left upper quadrant pain,  abdominal U/S was  non­ conclusive and  the  diagnosis of acute appendicitis was a long  shot,  however, on persistence of pain and increasing inflammatory parameters in blood exams aCT scan revealed intestinal mal-rotation with acute appendicitis.

of acute  abdomen; CT imaging  defined  20 cases  of gynecological emergencies which were misdiagnosed depending on  US examination. Moreover, CT imaging defined

3 cases of acute diverticulitis without abscess formation in two, but the 3rd was diverticular abscess. One case of cancer caecum forming mass mimicking appendicular mass  on US examination was diagnosed  on CT imaging. These data provide  an additional  advantage for preoperative CT which may modify  the surgical decision. In line with these findings, Purysko et al.20 documented that multidetector CT is an extremely useful noninvasive method for diagnosis and management of not only the most common causes of right lower quadrant abdominal pain such as appendicitis but also less  common conditions including inflammatory and infectious conditions involving the ileocecal region; diverticulitis; malignancies; conditions affecting the epiploic appendages, omentum,  and mesentery; and miscellaneous conditions.

It could be concluded that preoperative CT for patients with acute right lower abdominal pain reduces the negative surgical rate and improves true positive surgical  rate  and is mandatory especially in  suspicious cases. Moreover, preoperative CT could help differential diagnosis of the underlying pathology and so can modify surgical decision. Wider scale studies  are  mandatory for evaluation of  cost-benefit of  considering preoperative CT as a routine diagnostic procedure.

References:

1- Purkayastha S, Purkayastha S, Paraskevas P: Acute appendicitis: Weighing up risks and  benefits of investigations and treatments. BMJ 2006; 333(7569): 652-

653.

2- fives I, Paajanen HE, Herzig KH, Fagerstrom A, Miettinen  PJ: Changing incidence of acute appendicitis and  nonspecific abdominal pain between 1987 and 2007 in Finland. World J Surg 2011; 35(4): 731-

738.

3- Mayor R, Saenz NC, Kanegaye JT: Acute appendicitis presenting with  a painful