Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

Diagnosing Pneumoperitoneum with Abdominal Ultrasound

Three Part Question

In [an adult patient presenting with a possible perforated viscus] is [abdominal ultrasonography] able to rule in or rule out a [pneumoperitoneum]?

Clinical Scenario

A 42 year old man presents with severe constant upper abdominal pain with no nausea and vomiting preceded by two days of vague ‘heartburn’. He has been binge drinking over the weekend and is a smoker of 20 pack years. He had been diagnosed with peptic ulcer three years previously by his GP. He appears to be in obvious distress with a distended rigid abdomen, rebound tenderness, absent bowel sounds and is unwilling to sit up or be moved. His ECG is normal. You clinically suspect a perforated duodenal ulcer as your main differential diagnosis and as you resuscitate him you think if it is possible to detect free intra-peritoneal air using ultrasound.

Search Strategy

Ovid Medline

Database:
Ovid MEDLINE(R) In-Process & Other Non-Indexed Citations
Ovid MEDLINE(R) <1946 to May Week 4 2013>
Ovid OLDMEDLINE(R) <1946 to 1965>
Search Strategy:
--------------------------------------------------------------------------------
1 exp Pneumoperitoneum, Artificial/ or exp Pneumoperitoneum/ (5974)
2 exp Injections, Intraperitoneal/ (27864)
3 exp Air/ (20088)
4 extraluminal.mp. [mp=title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept, rare disease supplementary concept, unique identifier] (1375)
5 exp Uterine Perforation/ or exp Peptic Ulcer Perforation/ (6126)
6 perforation.mp. (47609)
7 1 or 2 or 3 or 4 or 5 or 6 (101773)
8 ultrasound.mp. [mp=title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept, rare disease supplementary concept, unique identifier] (143164)
9 exp Ultrasonography/ (236117)
10 Ultrasonographic.mp. [mp=title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept, rare disease supplementary concept, unique identifier] (16863)
11 Ultrasonography.mp. [mp=title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept, rare disease supplementary concept, unique identifier] (169566)
12 US.mp. (6689977)
13 Sonography.mp. (25932)
14 Sonographic.mp. (19735)
15 8 or 9 or 10 or 11 or 12 or 13 or 14 (6976916)
16 7 and 15 (26876)
17 limit 16 to (english language and humans) (8447)
18 limit 17 to \"diagnosis (best balance of sensitivity and specificity)\" (862)

Google Scholar (no time limit applied)
Searched for first 100 pages using the terms, ‘pneumoperitoneum’ and ‘ultrasound’

Search Outcome

Medline Ovid - A total of 862 articles were identified of which 5 was relevant for inclusion

Cochrane – No relevant articles

Google Scholar – 13,300 results of which 1 additional relevant article to Medline was found for inclusion

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Pinto et al
2000
Italy		Single centre over 9 months with 72 consecutive patients (25-72 years) referred with non-traumatic acute abdominal pain suspected of having viscus perforation examined in all 4 abdominal quadrants using a 3.5MHz convex probe and a 7.5MHz linear probeLevel 2b – Exploratory cohort study with good reference standards Screened for intra-peritoneal free gas looking for comet-tail or ring-down artifacts or the ‘shifting phenomenon’ when patient position changes All 63 patients who had surgery had gastrointestinal perforationPatients scanned by a senior radiologist. Sample size estimates not performed and no mention of ethical approval or independent gold standard blinding.
Secondary signs were looked for such as free intra-peritoneal fluid, thickening of bowel, gallbladder, gastric walls and local peritoneal inflammation9 patients had no perforations
Final diagnosis for all 72 patients confirmed with further radiographs, CT scans, endoscopy, laparotomy or clinical follow up41 gastro-duodenal perforations US 68.3% sensitivity for free intra-peritoneal air
22 intestinal perforations US 36.4% sensitivity for free intra-peritoneal air
Overall Sensitivity 57.14% Specificity 100.0% PPV 100% NPV 25%
Chen et al
2002a
Taiwan		Tertiary single centre study over 3 years of 132 patients (15-83 years) presenting to the emergency department with suspected hollow organ perforation prospectively analysed with US, upright chest radiograph, +/- left lateral decubitus abdominal radiography examinations. US performed by attending surgeon or emergency physician with a 3.75MHz curved-array transducer in supine and left lateral position over epigastrium and right hypochondrium Level 2b – Exploratory cohort study with good reference standards i) Direct US signs of perforation such as ring-down artifacts125 patients had laparotomy (121 had hollow organ perforation, 3 perforated appendicitis, 1 had acute cholecystitis)Calculations of sensitivity, specificity, PPV and NPV incorrect based on available data yet the conclusions of the paper would still be correct. Sample size estimates not performed and no mention of ethical approval or independent gold standard blinding.
ii) Indirect US signs of perforation such as free fluid, bowel dilatation, wall thicknessUS Sensitivity 92.6% Specificity 73.3% PPV 97% NPV 55% Accuracy 90.4%
iii) NormalPlain radiography Sensitivity 78.5% Specificity 73.3% PPV 96% NPV 30% Accuracy 77.9%
Final diagnosis for all 132 patients confirmed with further CT scans, endoscopy, laparotomy or clinical follow up
Chen et al
2002b
Taiwan		Tertiary single centre study over 4 years of 188 patients (15-88 years) presenting to the emergency department with suspected hollow organ perforation prospectively analysed with US, upright chest radiograph, +/- left lateral decubitus abdominal radiography examinations. US performed by attending surgeon or emergency physician with a 3.75MHz curved-array transducer in supine and left lateral position over epigastrium and right hypochondrium Level 2b – Exploratory comparative study with good reference standards i) Direct US signs of perforation such as echogenic lines and ring-down artifacts178 patients had laparotomy (170 had hollow organ perforation, 5 perforated appendicitis and 3 acute cholecystitis) Calculations of sensitivity, specificity, PPV and NPV incorrect based on available data yet the conclusions of the paper would still be correct. Sample size estimates not performed and no mention of ethical approval or independent gold standard blinding.
ii) Indirect US signs of perforation such as free fluid, bowel dilatation, wall thicknessUS Sensitivity 92.3% Specificity 55.6% PPV 95% NPV 43% Accuracy 88.8%
iii) NormalRadiographs Sensitivity 78.8%Specificity 55.6% PPV 94% NPV 22% Accuracy 76.6%
Final diagnosis for all 188 patients confirmed with further CT scans, panendoscopy, laparotomy or clinical follow up
Karahan et al
2004
Turkey		Single centre study over 32 months with 72 patients (6 days-79 years) with suspected gastro-intestinal tract perforation and examined using a 3.5MHz convex probe and a 7.5MHz linear probe in the supine and left lateral positionsLevel 2b – Exploratory comparative study with good reference standards Screened for intra-peritoneal free gas using the ‘scissors maneuver’ to demonstrate the acoustic reverberation or ring down artefact 28 patients had surgery and 22 confirmed to have gastrointestinal tract perforation surgically (16 open perforations and 6 sealed-off perforations). Remaining 6 had acute appendicitis (3) or negative laparotomy (3)US examinations performed by one senior radiologist. Sample size estimates not performed and no mention of ethical approval or independent gold standard blinding.
Final diagnosis for all 72 patients confirmed with further radiographs, laparotomy or clinical follow upUS Sensitivity 93.8% Specificity 100% PPV 100% NPV 98%
Radiographs Sensitivity 93.8% Specificity 100% PPV 100% NPV 98%
Asrani
2007
India 		Single centre prospective study of 600 consecutive patients (2-50 years) presenting to an emergency department with acute abdominal complaints. All patients had USS examinations performed with a 2.6-5MHz curved array transducer in 9 anatomical regions of the anterior abdomen.Level 1b – Validating diagnostic cohort study with good reference standardsThe presence or absence of the ‘Enhancement of the Peritoneal Stripe Sign’ (EPSS)600 patientsSingle person, single centre study. The author and investigator is a specialty radiologist. Each patient underwent a comprehensive US examination that lasted from 20-35 minutes. Side effects, specifically pain at the site during USS, not formally evaluated.
Other US findings incidentally noted EPSS positive in 24 patients EPSS negative in 576 patients
Final diagnosis for all 600 patients confirmed with further radiographs, CT scans, laparotomy or clinical follow up3 false positives 0 false negatives
Sensitivity 100% Specificity 99% PPV 87.5% NPV 100%
Moriwaki et al
2009
Japan		Single tertiary centre study over 5 years with 484 patients with severe chest-abdominal-pelvic blunt trauma or acute abdominal pain examined using a 3.5MHz convex probeLevel 2b – Exploratory cohort study with good reference standardsPrimary outcome – sensitivity and specificity of US in diagnosis of gastro-intestinal perforations looking for high-echoic area with a high echoic tail which move and change on compression54 patients diagnosed with gastro-intestinal perforationsSonographic examinations performed by experienced gastroenterologic or general surgeons with >5years US experience in fields other than emergency medicine and traumatology. Only the area of the liver was scanned. Sample size estimates not performed, no mention of ethical approval or blinding.
Final diagnosis confirmed by operative findings, CT scan, radiologic and/or clinical observation for >4 daysOverall Sensitivity 85.2%
Blunt trauma Sensitivity 85.7% Specificity 99.6%
Acute abdomen Sensitive 85.0% Specificity 100%

Comment(s)

The use of US to detect abnormal fluid in the abdominal cavity with the standard focused assessment with sonography for trauma (FAST) scan is well established in emergency medicine but the similar use of ultrasound to detect abnormal air in the abdominal cavity has lagged behind despite its ease, superiority and utility as a screening tool being well described since the pioneering work of Seitz and Reising (1982). This German paper reported a sensitivity of 90% and a specificity of 100% for the detection of pneumoperitoneum in 4000 consecutive patients with non-traumatic acute abdominal pain. US has been demonstrated since to detect pneumoperitoneum in restricted and experimental conditions (Nirapathpongporn et al, 1984; Lee et al, 1990; Chadha et al, 1993; Meuwly et al, 1993; Braccini et al, 1996; Grechenig et al, 1999) with Muradali et al (1999) confirming in animal and human studies the value of the ‘Enhancement of the Peritoneal Stripe Sign’ (EPSS) as a marker for intra-peritoneal air. Other sonographic markers of intra-peritoneal air have since been described such as the ‘Abdominal a-lines’ (Bomann et al, 2011) and the ‘Thickened Peritoneal Stripe’ sign in massive pneumoperitoneum (Bomann et al, 2011) which await clinical validation. The above Best Bet papers have been chosen as they reflect clinical practice in patients presenting to the emergency department beyond a mere case series but the results reflect the performance of experienced sonographers or surgeons who use ultrasound on a daily working basis. The poorer results from the Pinto et al (2000) paper may possibly be due to the less sensitive ring-down artefacts or the ‘shifting phenomenon’ (which requires a larger volume of gas) as the sole means of diagnosing intra-peritoneal free air instead of incorporating an additional ‘scissors maneuver’ (Karahan et al, 2004) or the EPSS (Asrani, 2007; Moriwaki et al, 2009). The skill-set involved in detecting abnormal free intra-peritoneal air is within the capability of emergency physicians who utilise ultrasound in their work and can be incorporated as an adjunct to their clinical practice to improve patient care where exposure to radiation (children, females of reproductive age and pregnancy) or re-positioning (critically ill and trauma) may be undesirable or impractical since evidence point to it being equivalent or superior to radiography or even equivalent to computed tomography in experienced hands (Hoffman et al, 2012).

Clinical Bottom Line

The EPSS is highly sensitive and specific for intra-peritoneal air with great utility as an adjunct for improving patient care but the subtle signs are patient and operator dependent, requiring practice, skill and experience for detection and interpretation. It should at present be regarded as a Rule-In and taken in context with other clinical findings until further validating work investigating its diagnostic utility by emergency physicians in a clinical setting is carried out.

References

  1. Pinto F, Scaglione M, Pinto A, Lassandro F, Romano L, Grasi R Gastrointestinal perforation: ultrasound diagnosis Emergency Radiology 2000;7(5):263-267
  2. Chen SC, Wang HP, Chen WJ, Lin FY, Hsu CY, Chang KJ, Chen WJ Selective use of ultrasonography for the detection of pneumoperitoneum Academic Emergency Medicine 2002;9(6):643-645
  3. Chen SC, Yen ZS, Wang HP, Lin FY, HSU CY, Chen WJ Ultrasonography is superior to plain radiography in the diagnosis of pneumoperitoneum The British Journal of Surgery 2002;89(3):351-354
  4. Karahan OI, Kurt A, Yikilmaz A, Kahriman G New method for the detection of intraperitoneal free air by sonography: scissors maneuver Journal of Clinical Ultrasound 2004;32(8):381-385
  5. Asrani A Sonographic diagnosis of pneumoperitoneum using the \\\'enhancement of the peritoneal stripe sign.\\\' A prospective study Emergency Radiology 2007; 14(1):29-39 Epub 2007 Mar 9
  6. Moriwaki Y, Sugiyama M, Toyoda H, Kosuge T, Arata S, Iwashita M, Tahara Y, Suzuki N Ultrasonography for the diagnosis of intraperitoneal free air in chest-abdominal-pelvic blunt trauma and criticial acute abdominal pain Archives of Surgery 2009;144(2):137-141; discussion 142
  7. Bomann JS, van Tonder R, Hernandez S, Moore C Abdominal a-lines: a simpler sonographic sign of pneumoperitoneum Critical Ultrasound Journal 2011;3(1)41-42
  8. Braccini G, Lamacchia M, Boraschi P, Bertellotti L, Marrucci A, Goletti O, Perri G Ultrasound versus plain film in the detection of pneumoperitoneum Abdominal Imaging 1996;21(5):404-412
  9. Chadha D, Kedar RP, Malde HM Sonographic detection of pneumoperitoneum: an experimental and clinical study Australasian Radiology 1993;37(2):182-185
  10. Grechenig W, Peicha G, Clement HG, Grechenig M Detection of pneumoperitoneum by ultrasound examination: an experimental and clinical study Injury 1999;30(3):173-178
  11. Hoffman B, Nürnberg D, Westergaard MC Focus on abnormal air: diagnostic ultrasonography for the acute abdomen European Journal of Emergency Medicine 2012;19(5):284-291
  12. Lee DH, Lim JH, Ko YT, Yoon Y Sonographic detection of pneumoperitoneum in patients with acute abdomen American Journal of Roentgenology 1990;154(1):107-109
  13. Meuwly JY, Fournier D, Hessler C, Schnyder PA Sonographic diagnosis of pneumoperitoneum in twelve patients European Radiology 1993;3(3):234-236
  14. Muradali D, Wilson S, Burns PN, Shapiro H, Hope-Simpson D A Specific Sign of Pneumoperitoneum on Sonography: Enhancement of the Peritoneal Stripe American Journal of Roentgenology 1999;173(5):1257-1262
  15. Nirapathpongporn S, Osatavanichvong K, Udompanich O, Pakanan P Pneumoperitoneum detected by ultrasound Radiology 1984;150(3):831-832
  16. Seitz K, Reising KD [Ultrasound detection of free air in the abdominal cavity] Ultraschall Ultraschall in der Medizin 1982;3(1)4-6