Assessing the constancy of intracranial aneurysm growth rates

Despite ongoing natural history studies,12 the natural history of unruptured cerebral aneurysms, including pat terns of growth, remains undefined. Elucidating this growth pattern may allow physicians to determine if spe­ cific periods of risk occur or if the risk to the patient is constant. Two possibilities exist for the pattern of aneu­ rysm growth: linear (constant) or episodic with periods of growth separated by periods without. Linear growth would theoretically be associated with a constant predictable risk based on the growth rate, which occasionally could be rapid,5,16 whereas episodic growth might be associat­ ed with periods of unpredictable increased or decreased risk. It has been assumed that cerebral aneurysms have linear growth, and the authors of the proceeding article have tested this hypothesis by evaluating the plausibil­ ity of a constant growth rate for intracranial aneurysms via math ematical modeling. They tested the plausibility of a constant growth rate by comparing hypothetical cohorts of patients with different initial mean constant lesion growth rates and the population­based incidence of sub arachnoid hemorrhage (SAH). The authors found that even within a hypothetical cohort with a growth rate that most closely resembled the incidence rates in the actual population, the specific incidence rates in the mod el differed substantially from those in the observed population data. They therefore concluded that the actual growth process must be episodic. This conclusion should be interpreted with caution as biological behavior is sel­ dom accurately predicted with mathematical modeling. An aneurysm is a complex active biological structure and not simply a balloon that obeys the LaPlace Law. Con stant biological changes are occurring at all times within the aneurysm wall and can be related to flow6 or molecular changes11 that are not clearly understood. Not all aneurysms are equal and some may indeed have con­ stant slow growth whereas others may be more unstable and associated with episodic growth based on factors that are currently unknown. Regarding the limitations of the paper by Koffijberg and colleagues, as with other mathematical modeling studies, assumptions are made that can hinder a con­ clu sive argument. This is particularly true if multiple as sump tions are made despite the authors’ arguments to the contrary. The first assumption to be addressed con­ cerns the multiplicity of aneurysms not influencing the results. Although the authors stated that the Internation­ al Study of Unruptured Intracranial Aneurysms did not show an increased risk of SAH with multiple aneurysms,3 which was also demonstrated by Juvela et al.,4 and therefore will not affect the results of their own study, it must be remembered that many other studies have shown that multiple aneurysms do increase the risk of SAH in patients. In 1974 Mount and Brisman7 reviewed 158 cases of unruptured multiple aneurysms and found a bleeding rate of at least 10% per year in patients with multiple aneurysms. Wiebers and associates13 have also noted that unruptured lesions that are part of multiple an eurysm constellations can have a greater propensity to rupture than solitary aneurysms, and this hypothesis was supported by data from Winn et al.14 The Finnish data also support the concept that patients with multiple aneu­ rysms have a higher likelihood of rupture of an unrup­ tured intracranial aneurysm. Heiskanen and Marttila2 have reported on 84 patients with multiple aneurysms in whom a ruptured lesion was definitely identified and treated at surgery. Of these 84 patients, 8 had a recur­ rent hemorrhage from the unruptured aneurysm during follow­up periods ranging from 4 months to 11 years with a rupture rate of > 1%/year. Heiskanen1 has also reported on a 10­year follow­up in 61 patients with unruptured intracranial aneurysms who had undergone surgery for a prior ruptured aneurysm and with a lesion rupture rate of > 1%/year. Japanese data also support an increased risk of hemorrhage in patients with multiple aneurysms, with Yasui and associates17 demonstrating an annual rupture rate of 6.8% for multiple aneurysms and 1.9% for single ones. Therefore, if the latter argument holds true, it would significantly change the results of the study by Koffijberg et al., as up to 30% of patients have multiple aneurysms. The second assumption that must be addressed is in re gards to ignoring the age­related mortality rate and its insignificance. This factor may not be as insignificant as the authors suggest. In the recent International Study of Unruptured Intracranial Aneurysms data,12 a patient’s age was especially important because, although it does not affect rupture rates, it has a substantial effect on sur gical morbidity and mortality rates, which would sig­ nificantly affect the results of their analysis. The third as sumption that needs to be addressed regards ignoring the association between growth and rupture risk. This fac tor is significant as growth does not always equal rup ture, and there are no direct clinical data that prove that growth really equates with rupture. Insufficient data