Logistic regression in case-control studies: The effect of using independent as dependent variables

In case-control studies, cases are sampled separately from controls. In such studies the primary analysis concerns the estimation of the effect of covariables on being a case or a control. To explore causal pathways, further secondary analysis could concern the relationships among the covariables. In this paper the validity of such secondary analysis is addressed. In particular, the use of multiple logistic regression in case-control studies where the dependent variable is not the case/control indicator is explored. It is shown that only under very restrictive conditions will sample regression coefficients correctly estimate their true value. In many situations, it may be valid to regress one covariable on others in the control group, but not in the case group or the combined sample. This principle is illustrated by a study of sexually transmitted disease in Kenya.     

The effects of pellicle formation on streptococcal adhesion to human enamel and artificial substrata with various surface free-energies

The influence of a pellicle on streptococcal adhesion was studied. A "ripened" two-hour salivary pellicle and an "early" five-minute salivary pellicle were formed on human enamel and artificial solid substrata with varying surface free-energies. Three strains of oral streptococci, also with widely different surface free-energies, were used for adhesion studies. Pellicle formation and streptococcal adhesion took place at a constant shear rate of 21 s-1. Adhesion of S. mitis BMS to bare and pellicle-covered enamel was low and not significantly affected by the presence of a pellicle (0.7 x 10(6) and 0.6 x 10(6) cells.cm-2, resp.), whereas the numbers of S. sanguis 12 and S. mutans NS adhering to bare enamel (4.2 x 10(6) cells.cm-2 and 13.8 x 10(6) cells, cm-2, resp) were significantly reduced by the presence of a pellicle. This reduction was almost complete after only five minutes of salivary protein adsorption (1.9 x 10(-6) and 1.1 x 10(6) cells.cm-2 for S. sanguis and S. mutans, resp.) but further reduced for S. sanguis adhering to a ripened pellicle (0.7 x 10(6) cells.cm-2). The numbers of streptococci adhering at equilibrium to bare enamel could be fitted to a thermodynamically based model, which was previously described for bacterial adhesion to homogenous artificial substrata. Streptococcal adhesion to artificial substrata exposed to saliva was low, and the differences among uncoated materials were markedly reduced even after only five minutes' exposure to saliva.     

Reply to 'Reciprocal translocation carriers ascertained for recurrent miscarriage have a possibility to yield an unbalanced fetal chromosome pattern'

Sir, We thank Mayumi Sugiura-Ogasawara et al. for their commentson our paper (Goddijn et al., 2004     

Skeletal tissue engineering-from in vitro studies to large animal models

Bone is a tissue with a strong regenerative potential. New strategies for tissue engineering of bone should therefore only focus on defects with a certain size that will not heal spontaneously. In the development of tissue-engineered constructs many variables may play a role, e.g. the source of the cells used, the design and mechanical properties of the scaffold and the concentration and mode of application of growth factor(s).Models for studying new strategies for tissue engineering of bone should be based on the target tissue to be restored. However, in light of the many potential variables, which may also interact if used in combination(s), there is also a large need for relatively simple models in which variables can be tested in a limited number of animals. Moreover, in compromised bone there may be a problem with the load-bearing capacity of the remaining healthy bone. In this light, an important prerequisite for tissue-engineering constructs is that they can be tested in loaded conditions. Particularly, this latter prerequisite is very difficult to achieve. Therefore, in vitro tests for mechanical stability are very useful for evaluating the mechanical consequences of a particular reconstruction procedure prior to the animal experiment. Before a tissue-engineered construct can be introduced into a clinical trial, a final test should be available in a large animal model that is as close and relevant to a particular problematic clinical situation as possible.In the past, a series of models were developed in our laboratory that are very useful for testing tissue-engineered constructs. In this paper, we focus on the use of relatively new simple in vitro and in vivo models for hip revision surgery, segmental bone defect restoration and tumour surgery.     

Experimental infection of immunomodulated NOD/LtSz-SCID mice as a new model for Plasmodium falciparum erythrocytic stages

The main objective of this study was to determine whether a chemical immunomodulation protocol could reduce the resistance of NOD/LtSz-SCID mice to Plasmodium falciparum infection and provide an improved mouse model for screening the antimalarial activity of new compounds. This model was compared with the presently used immunodeficient Beige/Nude/Xid (BNX) mouse model, using the same protocol, in terms of percentage of infected mice, parasite development, leukocyte response and phagocytosis of P. falciparum infected cells in various organs. Our results show that the combination of the chemical immune modulation protocol with the genetic background of NOD/LtSz-SCID mice results in the development of long-lasting P. falciparum infection in a high percentage of mice. A comparison of the results obtained in the histological study for both mouse models suggests that the higher rate of success in NOD/LtSz-SCID mice could be related to the reduced macrophage recruitment developed in different tissues to remove the parasite from blood.     

Erythropoietin as an angiogenic factor in gastric carcinoma

AIMS: Previous studies have shown that increased vascularity is associated with haematogenous metastasis and poor prognosis in gastric cancer. The role of erythropoietin (Epo) in angiogenesis has not been completely clarified, although its involvement has been reported. In this study we correlated microvascular density and Epo receptor (Epo-R) expression in endothelial and tumour cells with histopathological type in gastric cancer. METHODS AND RESULTS: Specimens of primary gastric adenocarcinomas obtained from 40 patients who had undergone curative gastrectomy were investigated immunohistochemically by using anti-CD31 and anti-Epo-R antibodies. Stage IV gastric carcinoma had a higher degree of vascularization than other stages, and Epo-R expression in both endothelial and tumour cells increased in parallel with malignancy grade and was highly correlated with the extent of angiogenesis. CONCLUSIONS: Epo-R level correlates with angiogenesis and progression of patients with gastric carcinoma and we suggest that Epo might have a trophic effect on the vasculature of the gastrointestinal tract. Understanding mechanisms of gastric cancer angiogenesis provides a basis for a rational approach to the development of an anti-angiogenic therapy in patients with gastric cancer.     

The thymus is a site of mast cell development in chicken embryos

Thymic mast cells were studied by light and transmission electron microscopy in chicken embryos during organogenesis. Mast cells made their first appearance at day 15. At days 16 and 17, there was a burst of mast cell development with a peak of 278 ± 54 cells/mm² at day 16. Then, mast cell density decreased until hatching. During the whole embryonic period, about 80% of mast cells localized to the thymic medulla. In the cortex, they were less numerous, and some rare mast cells could be identified in the capsule and septa. Thymic mast cells could be recognized in association with hematopoietic foci, but frequently they grew independently from areas of hematopoiesis and appeared as single cells interspersed among thymocytes, thymic epithelial cells, and interdigitating cells. They were often recognized in close relationship with the scanty and delicate extracellular matrix of the developing gland. Viewed by electron microscopy, mast cells were relatively small cells, with a few secretory granules. Exocytosis was never seen, but, notably, granules emptied in a piecemeal degranulation fashion. This study demonstrates that the chicken thymus is a site of mast cell development during embryogenesis. The high mast cell density we found suggests a possible role for these cells during thymus organogenesis.     

A possible role of tryptase in angiogenesis in the brain of mdx mouse, a model of Duchenne muscular dystrophy

Duchenne muscular dystrophy (DMD) is characterized by muscle degeneration and affects the CNS. Dystrophin is absent in muscle and CNS of both DMD patients and mdx mouse, a model of DMD. While the involvement of vascular compartment in DMD was poorly investigated, some studies suggested a role for mast cells (MC). Tryptase, contained in the MC granules, stimulates angiogenesis in vitro and in vivo. We demonstrated for the first time a correlation between the extent of angiogenesis and the number of tryptase-positive neurons and microvessels and suggest that the tryptase contained in the neurons and in the endothelial cells of the mdx mouse brain may be involved in the regulation of angiogenesis taking place in mdx mouse.     

In vivo imaging of apoptosis by 99mTc-Annexin V scintigraphy: visual analysis in relation to treatment response.

BACKGROUND AND PURPOSE: Anticancer therapy induces apoptosis in a dose- and time-dependent fashion. (99m)Tc-Hynic-rh-Annexin V scintigraphy (TAVS) enables non-invasive in vivo imaging of treatment-induced apoptosis. We identified the visual patterns of (99m)Tc-Hynic-rh-Annexin V tumour uptake and related these to treatment response. PATIENTS AND METHODS: Thirty-three patients with malignant lymphoma (n=26), leukaemia (n=1) NSCLC (n=5), H&NSCC (n=1), scheduled for radiotherapy (n=27), platinum-based chemotherapy (n=5) or concurrent chemoradiation (n=1), underwent TAVS before and early after the start of treatment. Planar and SPECT images were visually examined to assess changes in tumour (99m)Tc-Hynic-rh-Annexin V uptake. Twenty-nine patients were eligible for further analysis. Annexin V uptake before (U(baseline)) and early after (U(post)) the start of treatment was graded using a four-step scale: 0, absent; 1, weak; 2, moderate and 3, intense. The difference between these values (Delta U) was calculated and correlated to tumour response after therapy (Spearman rank correlation test). RESULTS: Weak to moderate U(baseline) was detected in 13/15 patients with a complete response and U(post) was markedly increased in all these cases (Delta U range 1-3). Partial response (n=7) was associated with weak to moderate U(baseline) and a moderately increased U(post) (Delta U range 1-2). In patients with stable disease (n=5), U(baseline) was predominantly weak, without considerable changes in uptake after the start of treatment (Delta U range 0-1). Finally, in case of progressive disease (n=2), either no tumour uptake or a decrease in U(post) was detected (Delta U=-1). A statistically significant correlation was found between changes in (99m)Tc-Hynic-rh-Annexin V tumour uptake and clinical response (correlation coefficient=0.62; P<0.001). CONCLUSIONS: Complete or partial tumour response was associated with a marked increase of (99m)Tc Hynic-rh-Annexin V accumulation early during treatment compared to baseline values. In case of stable or progressive disease, pretreatment scans demonstrated predominantly low (99m)Tc Hynic-rh-Annexin V tumour uptake and no significant increase early after treatment. These results indicate that TAVS might be useful as a predictive test for treatment response.