How Strongly Do Physical Examination Estimates and Ultrasonographic Measurements of Liver Size Correlate? A Prospective Study

Abstract

Background

Liver size assessed by physical examination and ultrasound has long been used to gain useful clinical information. The size measurements obtained by these modalities have been difficult to compare as they are measured in 2 different axes (transaxial vs midclavicular). Our objective was to identify a measurement correlation between ultrasound and physical examination liver size findings. We aimed to develop a correction factor whereby the liver size could be translated between the measured transaxial size obtained by ultrasound and physical examination size when measured in the midclavicular line.

Methods

We conducted a prospective study including 101 adult patients with liver disease undergoing liver biopsy between April 2008 and November 2008 at Penn State Health Milton S Hershey Medical Center. Liver measurements were obtained by physical examination and ultrasound, which were performed by a single hepatologist.

Results

The average physical examination size using the midclavicular approach was 8.9 cm ± 1.13. On ultrasound, the average transaxial measurement was 14.3 cm ± 1.6. A ratio was made between measurements from the midclavicular line physical examination size and transaxial ultrasound size, and found to have a mean correction factor of 1.6 ± 0.14. The correction factor was applied to the physical examination-determined liver size and compared with ultrasound findings, with 76% of values (77/101) falling within 10% of the ultrasound-determined liver size.

Conclusion

This study proves that a strong correlation exists between physical examination estimates of liver size and the measured size on ultrasonography. Multiplying the percussed liver span by a correction factor consistently yields accurate predictions of the transaxial liver span.

Introduction

Liver size assessed by physical examination has long been used to provide the clinician with useful information. Typically, the physical examination to assess liver size includes percussing the vertical span of the liver along the right midclavicular line during quiet respirations. Listening for a dull sound, the lower edge is percussed along the right costal margin, whereas the upper border is usually sharply defined by the air filled-lung.1, 2

A palpable liver edge below the right costal margin is often indicative of underlying liver disease.¹ Generally, the normal liver size is approximately 8 cm, with 6 cm considered small and greater than 10 cm considered hepatomegaly.³ More specifically, by midclavicular percussion, the mean liver size for females is 7 cm and for males is 10.5 cm, with a liver size 2 to 3 cm larger or smaller than these values being considered abnormal.⁴ Although practical and useful, comparing physical examination findings to more objective assessments such as imaging modalities or autopsy studies has shown that liver span obtained solely via right midclavicular line percussion may underestimate true liver size.⁵

Ultrasound (US) has long been a method used to assess liver pathology and is now the most widely used diagnostic technique to examine the liver.⁶ The transaxial or oblique US technique measures the liver from pole to pole and is commonly used to measure liver size. In terms of liver size determination, US can give a quantitative and reproducible estimate of the liver span.⁷ In addition to providing reliable measures of liver size, US may provide a valuable means of monitoring disease progression and response to treatments by tracking changes in the size and consistency of the liver.⁸ US has also been used as an adjunctive modality to aid in teaching first-year medical students how to conduct the abdominal examination.⁹ The low operating costs, absence of ionizing radiation, high patient tolerance, and ease of use make it an attractive modality to assess the liver. The information gathered from US has also been used to diagnose not just biliary obstructive processes, but also intrinsic liver disease.¹⁰ Riley and colleagues 10, 11 have found that sonographic characteristics alongside of physical examination could be used to predict the diagnosis of nonalcoholic fatty liver disease.

The use of both physical examination and US to evaluate the liver size remains an attractive strategy as both are easy and inexpensive to perform and both are widely available. However, in developing countries, there remains a significant lack of access to ultrasound and sonography education.¹² In resource-poor settings without access to valuable diagnostic tests such as ultrasonography, the physical examination may serve as the primary means of clinical information gathering. Therefore, a tool that correlates physical examination findings to what we expect to see on a more advanced diagnostic study such as ultrasound may be especially beneficial in these resource-poor settings.

A common misconception is the idea that the measurements obtained via physical examination and US are directly interchangeable. Since the two are measured in different planes, they cannot be directly compared. Therefore, the reported liver sizes can be a source of confusion when the US size documented does not match the physical examination findings.

To date, no study has been performed where liver size could be translated from the transaxial approach traditionally made by US, to the right midclavicular size obtained on physical examination and vice versa with the aid of a correction factor. The goal of this study was to determine if a good correlation exists and if so, to develop a correction factor whereby the liver size can be translated between the physical examination size when measured in the right midclavicular line and the transaxial size obtained by US.