A comparison of different bivariate correlated frailty models and estimation strategies

Frailty models are becoming increasingly popular in multivariate survival analysis. Shared frailty models in particular are often used despite their limitations. To overcome their disadvantages numerous correlated frailty models were established during the last decade. In the present study, we examine bivariate correlated frailty models, and especially the behavior of the parameter estimates when using different estimation strategies. We consider three different bivariate frailty models: the gamma model and two versions of the log-normal model. The traditional maximum likelihood procedure of parameter estimation in the gamma case with an explicit available likelihood function is compared with maximum likelihood methods based on numerical integration and a Bayesian approach using MCMC methods with the help of a comprehensive simulation study. We detected a strong dependence between the two parameter estimates (variance and correlation of frailties) in the bivariate correlated frailty model and analyzed this dependence in detail. [Authors]]]> Bayes Theorem ; Likelihood Functions ; Models, Statistical ; Multivariate Analysis ; Survival Analysis oai:serval.unil.ch:BIB_7E89B55E49D8 2022-05-07T01:21:20Z <oai_dc:dc xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd"> https://serval.unil.ch/notice/serval:BIB_7E89B55E49D8 (reply) An occlusive or toxic mechanism affecting the vascular bed of short posterior ciliary arteries possibly underlies the sectoral choroidal atrophy after superselective ophtalmic artery chemotherapy info:doi:10.1097/IAE.0b013e3182265198 info:eu-repo/semantics/altIdentifier/doi/10.1097/IAE.0b013e3182265198 Munier, Francis Beck-Popovic, Maja Balmer, Aubin Binaghi, Stefano Bovey, Etienne Gaillard, Marie-Claire info:eu-repo/semantics/article article 2011-09-01 Retina, vol. 31, no. 8, pp. 1747-1748 info:eu-repo/semantics/altIdentifier/pissn/0275-004X <![CDATA[To the Editor: We thank Dr. Abramson et al for their interesting comments regarding our article.1 Abramson et al specifically challenge that choroidal vascular occlusion is the sole mechanism responsible for the fundoscopic changes illustrated in Figures 1 (Patient 1) and 2 (Patient 9), but rather think that the pattern of the chorioretinal lesions is related to the area of previous retinal detachment. They postulate a subretinal accumulation of Melphalan leading to retinal pigment epithelium and retinal toxicity. However, the proposed mechanism not only does not account for the sharply demarcated choriocapillaris occlusion (see Figure 1F in Munier et al 1) but also fails to correlate with previously detached areas. In support of this we show a novel fundus montage from Patient 1 (Figure 1) 3 weeks after the first superselective ophthalmic artery chemotherapy (SOAC) session. The extent of the retinal detachment is well delineated and almost symmetrically opposed to the final occluded chorioretinal territory (see Figure 1F in Munier et al 1). Similarly Patient 9 presented a total retinal detachment after the first SOAC session, which spontaneously reattached except inferonasally (Figure 2). Here again there is no correlation between retinal detachment and choriocapillaris atrophy. The fact that large choroidal vessels are still angiographically visible in the atrophic area in Patient 1 (see Figure 1F in Munier et al 1) is not inconsistent with the occlusion of the temporal short posterior ciliary arteries leading to choriocapillaris infarction, eventually followed in this case by spontaneous reperfusion of the large vessels. This interpretation is further supported by Figure 2 from Patient 9, in whom the altered vascular bed is shown to spare the far peripheral choroid normally perfused by peripheral retrograde choroidal arteries originating from the long posterior ciliary arteries and/or the anterior ciliary arteries.2 Alternatively it is possible that choroidal (and not subretinal) accumulation of Melphalan, secondary to an uneven drug distribution within the ophthalmic artery leading to preferential temporal short posterior ciliary perfusion of Melphalan, may cause choriocapillaris and retinal pigment epithelium toxic infarction. According to our physiopathologic hypotheses, the sectoral aspect of the chorioretinal atrophy would be the result of confluent triangular syndromes of occlusive or toxic etiology. Fig. 1 opens a modal window Fig. 1 Fig. 2 opens a modal window Fig. 2 Finally we agree with Abramson et al that an alteration of the vascular pressure during the SOAC procedure may well represent an alternative etiology of the Roth spots in Patient 3. In conclusion, we agree that SOAC appears to be highly efficient for advanced retinoblastoma. However, we think that safety and visual prognosis associated with SOAC were not paid the attention they deserve. Our study aimed at safety issues is the first to draw attention to visual outcome. The recent literature contains an increasing number of reports with positive SOAC results, while the negative ones tend to remain unpublished. We know from floor discussions during specialty meetings that several groups experienced severe complications such as central retinal artery occlusions (all unpublished, but Shields and Shields 3). Since the publication of our article, we observed an additional case of temporal chorioretinal atrophy and light perception in a patient seeking a second opinion after SOAC performed elsewhere. In this case, fluorescein angiography, performed 3 months after the third injection, displayed a silent choroid, that is, complete choroid occlusion and obviously no reperfusion of the large vessels (data not shown). We therefore remind that, without further evidence regarding safety and final vision, SOAC indications should be restricted to advanced retinoblastoma as an alternative to external beam radiotherapy and/or enucleation