Study Population

We conducted a retrospective cohort study of patients with cirrhosis diagnosed with HCC at Parkland Memorial Health & Hospital System, the safety net system for Dallas County, between January 2005 and June 2012. With 11 primary care clinics in low-income neighborhoods, Parkland cares for approximately 50% of patients with HCC in Dallas County. Given its integrated structure and function as the safety net institution of Dallas, patients admitted to Parkland Hospital often receive their continuity care through the Parkland Health System clinics. Parkland adopted a comprehensive electronic medical record (EMR), including laboratory and radiology results, for inpatient and outpatient care in 2008. Patients were identified by ICD-9 codes for HCC (155.0 or 155.2), tumor conference presentation lists, and databases of patients who underwent surgical (resection or transplantation) or interventional (transarterial chemoembolization [TACE] or local ablation) treatments for HCC as previously described.¹⁴

Two authors (A.G.S. and A.C.Y.) adjudicated HCC cases to confirm that they met diagnostic criteria based on American Association for the Study of Liver Diseases guidelines.² For tumors larger than 1 cm, diagnosis was made by a typical vascular pattern on dynamic imaging (arterial enhancement and delayed washout) or histology. This study was approved by the Institutional Review Board of UT Southwestern Medical Center.

Data Collection

Patient demographics, clinical history, laboratory data, and imaging results were obtained through review of EMRs and paper medical records. Two investigators (A.G.S. and N.P.) independently extracted information using standardized forms, with discrepancies resolved through discussion and consensus. Age, sex, race/ethnicity, and lifetime alcohol and smoking history were recorded, with active alcohol abuse defined as drinking more than 40 g/d. Date of first medical encounter and number of primary care and hepatology clinic visits were documented. Data regarding liver disease included underlying etiology and presence of decompensation (ascites or encephalopathy). We classified patients according to liver disease etiology, including HCV, hepatitis B virus, alcohol-related liver disease, NAFLD, and other. NAFLD was often associated with components of the metabolic syndrome (obesity, diabetes, dyslipidemia) but was a diagnosis of exclusion, only made in the absence of other causes of liver disease, including viral hepatitis and alcohol abuse. Laboratory data of interest included platelet count, creatinine, aspartate aminotransferase, alanine aminotransferase, bilirubin, albumin, international normalized ratio, and alpha fetoprotein (AFP) at presentation and diagnosis.

Dates of HCC presentation, diagnosis, and treatment were abstracted. The date of HCC presentation was defined by the presence of any suspicious liver mass on any imaging study, an AFP level greater than 100 ng/mL, or 2 consecutive increasing AFP levels greater than 20 ng/mL. We included a one-time AFP level of 100 ng/mL based on its greater specificity and associated higher level of clinical concern for HCC at this cutoff. For cases with consecutive increasing AFP levels, we defined presentation using the date of the second AFP test. The date of HCC diagnosis was defined as the date of characteristic imaging (as described earlier) or biopsy. Tumor characteristics were determined by imaging studies, which had all been interpreted by radiologists at our institution, and tumor staging was performed using the Barcelona Clinic Liver Cancer (BCLC) staging system. HCC treatments were categorized as liver transplantation, resection, radiofrequency ablation (RFA), TACE, systemic chemotherapy, or best supportive care. In patients who received multiple treatments, treatment was categorized as the most curative (ie, liver transplantation > resection > RFA > TACE > systemic chemotherapy > best supportive care). All patients with HCC at Parkland are evaluated in a multidisciplinary conference and clinic for evaluation and treatment.¹⁵

Statistical Analysis

Our primary outcome of interest was time from presentation to diagnosis. We used Kaplan-Meier analysis to determine time to diagnosis and Cox multivariate regression analysis to identify factors associated with time to diagnosis. We also analyzed time to diagnosis as a dichotomous outcome, with delayed diagnosis being defined as time from presentation to diagnosis exceeding 3 months (ie, 90 days). The cutoff of 3 months was based on prior studies examining delays in HCC care and tumor doubling time.¹³ Demographics, clinical features, and tumor characteristics were compared between the 2 groups using Fisher exact and Mann-Whitney rank sum tests for categorical and continuous variables, respectively. We assessed patient sociodemographic and clinical characteristics, including age, sex, race/ethnicity, primary language, alcohol abuse, insurance status, performance status, inpatient versus outpatient status, number of primary care visits, receipt of hepatology subspecialty care, cause of liver disease, platelet count, bilirubin, Child-Pugh class, and implementation of the EMR as independent variables. Statistical significance was defined as P less than .05 for both univariate and multivariate analysis. All data analysis was performed using Stata 11 (StataCorp, College Station, TX).

Patient Characteristics

Between January 2005 and July 2012, 457 patients with cirrhosis were diagnosed with HCC. Baseline characteristics of the cohort are shown in Table 1. The median age of patients was 56 years (range, 28–92 years), and more than 75% were men. Our population was racially diverse, with 36% African Americans, 30% Hispanic Caucasians, and 26% non-Hispanic Caucasians. Nearly 56% of patients were uninsured, 18% had Medicaid, and only 25% had Medicare or private health insurance. The most common causes of cirrhosis were HCV infection (69%), alcohol-induced liver disease (12%), and NAFLD (9%). The median Child-Pugh score at time of HCC diagnosis was 7 (range, 5–15), with 37% of patients having Child-Pugh A cirrhosis. Tumor stage at presentation was diverse, with 29% BCLC stage A, 9% BCLC stage B, 29% BCLC stage C, and 33% BCLC stage D.

Presentation Data

Most patients presented initially with an abnormal imaging study, with 219 (48%) presenting with abnormal abdominal ultrasound and 149 (33%) presenting with a mass on CT or MRI. AFP level was elevated in 135 patients (30%) including 67 whose presentation was an elevated AFP level alone (ie, without concomitant abnormal imaging). The other 22 patients presented to an outside hospital and the test at presentation was unknown given the lack of details in the available medical records. Nearly 40% (n=178) presented before the EMR system was implemented, whereas the other 279 presented after its adoption. Initial presentation of HCC occurred as an inpatient in 243 patients (53%), of whom 231 (Group 1) had the diagnostic process completed during the same hospitalization before discharge. The remaining 226 patients (Group 2) either presented or were diagnosed as an outpatient (Figure 1).

Time From Presentation to Diagnosis

The median time from presentation to diagnosis among all patients was 0.6 months. More than 90% (n=214) of patients in Group 1 (ie, inpatients) had a time from presentation to diagnosis of less than 1 week. Only 1 of these 231 patients had HCC diagnosed more than 1 month after initial presentation (ie, 1.5 months). The median time to diagnosis among the 226 patients in Group 2 was 2.2 months, with 87 (39%) being diagnosed more than 3 months after initial presentation.

Overall, 87 patients (39%) were diagnosed more than 3 months after initial presentation. Among the patients with diagnostic delay, 73 (84%) had at least one interval clinic visit to primary care or gastroenterology clinics, during which HCC presentation was not recognized. In fact, 52 patients (60%) had more than 3 visits and 20 (23%) had 10 or more clinic visits before diagnostic workup was completed.

Table 1

Patient Characteristics

View Table

View Full Size

Patient cohorts.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 13, 5; 10.6004/jnccn.2015.0074

An infiltrative type pattern was seen in 38 patients on follow-up imaging, and therefore interval growth was difficult to accurately determine. Among the 49 patients with diagnostic delays and massforming HCC, 9 (18%) were noted to have interval HCC growth of 2 cm or more in maximum diameter. However, no significant difference was seen in receipt of treatment (P=.68) between patients with and without diagnostic delays after adjusting for BCLC stage and presentation as an inpatient. Similarly, we did not detect a difference in survival according to the presence of diagnostic delay (P=.16) after adjusting for BCLC stage, presentation as an inpatient, and receipt of HCC-directed treatment.

Predictors for Diagnostic Delays

Diagnostic delays among all patients were positively associated with the presence of hepatic encephalopathy (odds ratio [OR], 3.39; 95% CI, 1.72–6.68) and negatively associated with presentation as an inpatient (OR, 0.17; 95% CI, 0.09–0.30), presentation after EMR implementation (OR, 0.33; 95% CI, 0.19–0.57), and presentation with an abnormal ultrasound (OR, 0.39; 95% CI, 0.22–0.67). Higher rates of diagnostic delays were observed among those with hepatic encephalopathy (31% vs 17%), whereas lower rates were seen in those who presented as an inpatient (9% vs 30%), presented after EMR implementation (13% vs 29%), and those who presented with an abnormal ultrasound with or without elevated AFP (26% vs 12%). Rates of diagnostic delays were 12.3% among those who presented with any abnormal ultrasound, compared with 29.9% for those who presented with an abnormal AFP and 24.8% among those who had an incidental mass discovered on CT/MRI.

Among the subset of patients in Group 2, diagnostic delays were positively associated with the presence of hepatic encephalopathy (OR, 2.29; 95% CI, 1.03–5.07) and negatively associated with presentation after EMR implementation (OR, 0.28; 95% CI, 0.15–0.52) and presentation with an abnormal ultrasound (OR, 0.36; 95% CI, 0.19–0.67). Higher rates of diagnostic delays were observed among those with hepatic encephalopathy (56% vs 35%), whereas lower rates were seen among those who presented after EMR implementation (26% vs 60%) and those who presented with an abnormal ultrasound with or without elevated AFP levels (27% vs 50%). Sex and presentation to primary care or gastroenterology clinic were significant on univariate analysis but failed to reach statistical significance on multivariate analysis.