Article Text

Download PDFPDF
PWE-131 Bone mineral density: a poor predictor of fracture risk in cirrhosis?
  1. L Chinnappan1,
  2. F Yousuf2,
  3. A Begum1,
  4. P Richardson1,
  5. I Patanwala1
  1. 1Gastroenterology, Royal Liverpool University Hospital NHS Trust, Liverpool
  2. 2Gastroenterology, Royal Gwent Hospital, Newport, UK


Introduction Metabolic bone disease is a recognised complication of cirrhosis and results in an increased incidence of fractures which incur significant morbidity in these patients. Dual energy x-ray absorptiometry (dexa) scans are recommended on a 2-yearly basis to screen for osteoporosis based on T scores along with calcium and vitamin D supplementation in patients with cirrhosis or severe cholestasis. Treatment with bisphosphonates is recommended to reduce the risk of fracture in those who are found to be osteoporotic.

Aim The aim of this study was to assess the role of bone mineral density measurements in stratifying fracture risk in patients with cirrhosis.

Method Cirrhotic patients who had undergone a DEXA scan between 2009 and 2013 were identified. Demographic data, liver disease stage (MELD and Chil Pugh Score), T scores as obtained at DEXA scan and fracture occurrence were recorded. Multivariate analyses was carried out using variables that were found to be statistically significant on Univariate analysis.

Results 79 cirrhotic patients with a mean age of 58(±10) yrs, 45% of whom were female were identified for the study and followed up for a median of 33 months (IQR 17–51). The aetiology of liver disease was alcohol (49%), autoimmune liver disease (AIH/PBC) (19%), NASH (15%), HCV (8%) and others (19%). Median MELD and CP Score for the cohort was 10 and 6 respectively. Normal mineral bone density, osteopenia and osteoporosis were found in 27%, 58% and 15% patients respectively. Fractures occurred in 24% (95% CI 15–35) of the entire cohort and in 19%, 24% and 33% of patients with normal BMD, osteopaenia and osteoporosis. Aetiology (including alcohol) was not associated fracture risk (p = 0.846). On univariate analysis only MELD and Child Pugh Score were associated with fractures (see Table 1below). There was no statistical difference in the T Score in patients with or without fractures. On multivariate analysis only Child Pugh score was independently associated with fracture risk (p = 0.024).

Abstract PWE-131 Table 1

Univariate analysis of variables in cirrhotic patients who fracture and those are fracture free

Conclusion Fractures occur in a significant proportion of patients with cirrhosis. DEXA scans to assess bone mineral density may not be sufficient in predicting the risk of fracture in this cohort. The severity of liver disease as defined by the Child Pugh score may be better at identifying patients at a higher risk in whom preventative strategies should be focused.

Disclosure of interest None Declared.


  1. J D Collier, M Ninkovic, J E Compston. Guidelines on the management of osteoporosis associated with chronic liver disease. Gut. 2002;50:suppl 1 i1-i9 doi:10.1136/gut.50.suppl_1.i1

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.