In a prospective study, 179 adult women (age range, 17-82 years) were followed up for 12 months after an index episode of community-acquired cystitis caused by Escherichia coli. Episodes of symptomatic urinary tract infection (UTI) were recorded, and urinary isolates were compared with the index episode isolate; 147 UTI episodes were detected during the follow-up. Of these episodes, 131 were classified as recurrences occurring at least 1 month after the index episode; 44% of the patients had recurrences. A history of UTI increased the risk of recurrence; only 11.8% of the 17 patients without previous episodes of UTI had at least one recurrence, while 47.5% of those with previous episodes had at least one recurrence (OR, 6.8; univariate logistic regression). E. coli caused 78% of the recurrent episodes. Phenotypic and genotypic analysis of E. coli strains showed that one-third of the recurrences were caused by the index episode strain, which could persist and cause recurrences throughout the 1-year follow-up period. The prevalence of adhesins or other identified virulence factors for UTI among the recurrence strains was identical to that among the index episode strains. The presence of these factors did not affect the risk of recurrence but did increase the likelihood that the index episode strain would persist and cause recurrent episodes of UTI.

Objective: HA (hyaluronan) is involved in cell migration, differentiation and cell proliferation, which all are essential to tumour growth. In addition, the cell surface receptor of HA, CD44, is important in cancer cell adhesion, cell migration and tumour neovascularisation. We studied the expression of HA and CD44 and their relationship with other prognostic factors and prostate-specific antigen (PSA) recurrence in local prostate cancer (PC). Material and Methods: 77 PC patients treated with radical prostatectomy were followed-up for a mean of 4 years. HA was detected by using a HA specific probe and CD44 expression was analysed by conventional immunohistochemistry. Results: All specimens expressed HA in tumour stroma and 78% (60/77) of the tumours showed strong stromal expression of HA. The fraction of positively stained specimens for CD44 was 66% (51/77). The strong stromal HA expression was related to perineural infiltration (p = 0.001) and capsule invasion (p = 0.05). No correlation was demonstrated between the stromal HA expression and CD44 expression, preoperative PSA, clinical or pathological T classification, pN status, Gleason grade, seminal vesicle invasion or surgical margin invasion. Reduced CD44 expression was related only to preoperative PSA level (p = 0.008). The PSA recurrence was predicted by strong stromal HA expression, pT classification, seminal vesicle invasion, capsule invasion and surgical margin invasion (p ≤ 0.02 for all). pT classification and seminal vesicle infiltration predicted PSA recurrence free survival (RFS). In the multivariate analysis only the seminal vesicle infiltration was an independent predictor of PSA RFS (p = 0.009). Conclusion: The strong stromal HA expression was related to PSA recurrence, perineural infiltration and seminal vesicle invasion. Those are the factors related to unfavourable prognosis, but the follow-up period of this series is too short to make definitive prognostic conclusions.

Summary A retrospective analysis of 321 gastric cancer patients was made to assess the prognostic value of TNM classification, tumour differentiation, Lauren classification, proliferative rate, inflammatory reaction and tumour invasion in vascular or neural structures of the gastric wall. The TNM classification showed the strongest correlation with survival in univariate and multivariate analyses (P<0.0001). The invasion in lymphatic or vascular system and Lauren classification were also independent prognosticators in multivariate analysis (P<0.05). In univariate analysis, the WHO-grade, the size and the location of the tumour and perinueral invasion were significant prognostic factors (P<0.01), as were the infiltration of lymphocytes and plasma cells in the tumour (P<0.05). On the other hand, the mitotic indices reflecting the proliferative activity of the tumour cells showed no significant correlation with the prognosis. The results indicate that the prognostic power of the TNM classification can be further increased by assessment of the above special histological features in gastric cancer.

A consecutive series of breast carcinomas (n = 595) followed up for a minimum of 7.5 years was analyzed for clinical, histological and morphometric prognostic factors after 5-years’ follow-up. Tumor size, nodal status, mitotic frequency and patients1 age at diagnosis predicted survival at 7.5 years highly significantly (p < 0.0001). Tubule formation (p = 0.002), histological grade (p = 0.026) and nuclear pleomorphism (p = 0.046) were related to prognosis as well. In Inpatients mitotic frequency (p = 0.018) was the best predictor of survival at 7.5 years whereas in N+ tumors tumor size (p = 0.0013), tubule formation (p = 0.003) and mitotic frequency (p = 0.0049) were the best predictors of survival at 7.5 years. In univariate survival analysis the age of the patients (p = 0.002), tubule formation (p = 0.005), axillary lymph node status (p = 0.008), lymphocyte infiltration (p = 0.049), mitotic frequency (p = 0.06) and tumor size (p = 0.08) predicted survival after 5 years’ follow-up. Tubule formation predicted survival (p = 0.025) in N – patients and in N + tumors the age of the patients was the most important predictor (p = 0.006). In Cox’s analysis tubule formation (p = 0.13), axillary lymph node status (p = 0.033), SD of nuclear perimetry (p = 0.046) and intraductal growth (p = 0.059) predicted survival independently after 5 years’ follow-up.

The biopsy specimens from the primary tumors of 234 women with axillary-lymph-node-positive breast carcinomas (followed up for a mean of 10.9 years) were subjected to interactive morphometric analysis of nine nuclear factors. The proliferative activity of the tumors was estimated by determining two different mitotic indices. Morphometrically determined nuclear factors and mitotic indices showed a significant correlation to the histological grading (p < 0.0001). Mitotic activity index (MAI; p = 0.018) and volume-corrected mitotic index (M/V index; p = 0.005) accurately predicted the tumor recurrence. Recurrence-free survival was related to the M/V index (p = 0.0003), MAI (p = 0.0024) and tumor size (p = 0.0144). Disease-related survival was determined by the tumor size (p < 0.0001), M/V index (p = 0.0142) and MAI (p = 0.0492) in that order. On the other hand, the nuclear factors analyzed and the histological grading used had no predictive value (i.e. tumor recurrence, recurrence-free survival or tumor-related survival) in these women. The results indicate that mitotic indices can be sucessfully applied in place for subjective grading and nuclear morphometry in predicting the disease outcome in patients with axillary-lymph-node-positive breast carcinomas. The mitotic indices provide independent prognostic information in addition to tumor size. The major clinical implications of these results would be to accurately disclose among these women the high-risk patients (i.e. those with high mitotic indices), who might benefit from more agressive adjuvant therapies.