Demetris et al used histologic features in a biopsy from a liver transplant to predict an unfavorable outcome. The authors are from the University of Pittsburgh, Mayo Clinic NYC, University of California San Francisco, and the University of Nebraska.
Models:
(1) short-term prognosis
(2) long-term prognosis
Unfavorable short-term outcome consists of one or more of the following:
(1) acute rejection ending in graft failure
(2) OKT3 or ALG used to treat the rejection
(3) any secondary treatment required
(4) complete resolution of the episode was not achieved within 21 days
Unfavorable long-term outcome consists of one or both of the following:
(1) rejection-related death within 6 months of rejection onset
(2) need for retransplantation within 6 months of rejection onset
Parameters:
(1) intensity of the portal tract inflammation
(2) lobular disarray and ballooning of hepatocytes
(3) cholestasis
Parameter |
Finding |
Points for Short Term Model |
Points for Long Term Model |
portal inflammation |
none |
0 |
0 |
|
mild |
0 |
0 |
|
moderate |
0.78 |
1.52 |
|
severe |
0 |
4.54 |
lobular disarray and ballooning |
none |
0 |
0 |
|
mild |
0.71 |
0 |
|
moderate |
-0.58 |
0 |
|
severe |
0 |
0 |
cholestasis |
absent |
0 |
0 |
|
present |
-0.86 |
0 |
value of X for short-term outcome
= SUM(points for short-term parameters) - 0.31
probability of an unfavorable short-term outcome =
= 1 / (1 + EXP((-1) * X))
value of Y for long-term outcome =
= SUM(points for long-term parameters) - 3.85
probability of an unfavorable long-term outcome =
= 1 / (1 + EXP((-1) * X))
Specialty: Gastroenterology, Surgery, general