Correcting mortality for loss to follow-up: a nomogram applied to antiretroviral treatment programmes in sub-Saharan Africa

Background The World Health Organization estimates that in sub-Saharan Africa about 4 million HIV-infected patients had started antiretroviral therapy (ART) by the end of 2008. Loss of patients to follow-up and care is an important problem for treatment programmes in this region. As mortality is high in these patients compared to patients remaining in care, ART programmes with high rates of loss to follow-up may substantially underestimate mortality of all patients starting ART. Methods and Findings We developed a nomogram to correct mortality estimates for loss to follow-up, based on the fact that mortality of all patients starting ART in a treatment programme is a weighted average of mortality among patients lost to follow-up and patients remaining in care. The nomogram gives a correction factor based on the percentage of patients lost to follow-up at a given point in time, and the estimated ratio of mortality between patients lost and not lost to follow-up. The mortality observed among patients retained in care is then multiplied by the correction factor to obtain an estimate of programme-level mortality that takes all deaths into account. A web calculator directly calculates the corrected, programme-level mortality with 95% confidence intervals (CIs). We applied the method to 11 ART programmes in sub-Saharan Africa. Patients retained in care had a mortality at 1 year of 1.4% to 12.0%; loss to follow-up ranged from 2.8% to 28.7%; and the correction factor from 1.2 to 8.0. The absolute difference between uncorrected and corrected mortality at 1 year ranged from 1.6% to 9.8%, and was above 5% in four programmes. The largest difference in mortality was in a programme with 28.7% of patients lost to follow-up at 1 year. Conclusions The amount of bias in mortality estimates can be large in ART programmes with substantial loss to follow-up. Programmes should routinely report mortality among patients retained in care and the proportion of patients lost. A simple nomogram can then be used to estimate mortality among all patients who started ART, for a range of plausible mortality rates among patients lost to follow-up.

Characterization of Optically Stimulated Luminescent Detectors in Photon & Proton Beams for Use in Anthropomorphic Phantoms

This study investigated characteristics of optically stimulated luminescent detectors (OSLDs) in protons, allowing comparison to thermoluminescent detectors, and to be implemented into the Radiological Physics Center’s (RPC) remote audit quality assurance program for protons, and for remote anthropomorphic phantom irradiations. The OSLDs used were aluminum oxide (Al2O3:C) nanoDots from Landauer, Inc. (Glenwood, Ill.) measuring 10x10x2 mm3. A square, 20(L)x20(W)x0.5(H) cm3 piece of solid water was fabricated with pockets to allow OSLDs and TLDs to be irradiated simultaneously and perpendicular to the beam. Irradiations were performed at 5cm depth in photons, and in the center of a 10 cm SOBP in a 200MeV proton beam. Additionally, the Radiological Physics Center’s anthropomorphic pelvic phantom was used to test the angular dependence of OSLDs in photons and protons. A cylindrical insert in the phantom allows the dosimeters to be rotated to any angle with a fixed gantry angle. OSLDs were irradiated at 12 angles between 0 and 360 degrees. The OSLDs were read out with a MicroStar reader from Landauer, Inc. Dose response indicates that at angles where the dosimeter is near parallel with the radiation beam response is reduced slightly. Measurements in proton beams do not show significant angular dependence. Post-irradiation fading of OSLDs was studied in proton beams to determine if the fading was different than that of photons. The fading results showed no significant difference from results in photon beams. OSLDs and TLDs are comparable within 3% in photon beams and a correction factor can be posited for proton beams. With angular dependence characteristics defined, OSLDs can be implemented into multiple-field treatment plans in photons and protons and used in the RPC’s quality assurance program.

DEVELOPMENT AND IMPLEMENTATION OF A REMOTE AUDIT TOOL FOR HIGH DOSE RATE (HDR) 192IR BRACHYTHERAPY USING OPTICALLY STIMULATED LUMINESCENCE DOSIMETRY

This work aimed to create a mailable and OSLD-based phantom with accuracy suitable for RPC audits of HDR brachytherapy sources at institutions participating in NCI-funded cooperative clinical trials. An 8 × 8 × 10 cm3 prototype with two slots capable of holding nanoDot Al2O3:C OSL dosimeters (Landauer, Glenwood, IL) was designed and built. The phantom has a single channel capable of accepting all 192Ir HDR brachytherapy sources in current clinical use in the United States. Irradiations were performed with an 192Ir HDR source to determine correction factors for linearity with dose, dose rate, and the combined effect of irradiation energy and phantom construction. The uncertainties introduced by source positioning in the phantom and timer resolution limitations were also investigated. It was found that the linearity correction factor was where dose is in cGy, which differed from that determined by the RPC for the same batch of dosimeters under 60Co irradiation. There was no significant dose rate effect. Separate energy+block correction factors were determined for both models of 192Ir sources currently in clinical use and these vendor-specific correction factors differed by almost 2.6%. For Nucletron sources, this correction factor was 1.026±0.004 (99% Confidence Interval) and for Varian sources it was 1.000±0.007 (99% CI). Reasonable deviations in source positioning within the phantom and the limited resolution of the source timer had insignificant effects on the ability to measure dose. Overall measurement uncertainty of the system was estimated to be ±2.5% for both Nucletron and Varian source audits (95% CI). This uncertainty was sufficient to establish a ±5% acceptance criterion for source strength audits under a formal RPC audit program. Trial audits of eight participating institutions resulted in an average RPC-to-institution dose ratio of 1.000 with a standard deviation of 0.011.

Development and Implementation of the Use of Optically Stimulated Luminescent Detectors in the Radiological Physics Center Anthropomorphic Quality Assurance Phantoms

The Radiological Physics Center (RPC) uses both on-site and remote reviews to credential institutions for participation in clinical trials. Anthropomorphic quality assurance (QA) phantoms are one tool the RPC uses to remotely audit institutions, which include thermoluminescent dosimeters (TLDs) and radiochromic film. The RPC desires to switch from TLD as the absolute dosimeter in the phantoms, to optically stimulated luminescent dosimeters (OSLDs), but a problem lies in the angular dependence exhibited by the OSLD. The purpose of this study was to characterize the angular dependence of OSLD and establish a correction factor if necessary, to provide accurate dosimetric measurements as a replacement for TLD in the QA phantoms. A 10 cm diameter high-impact polystyrene spherical phantom was designed and constructed to hold an OSLD to study the angular response of the dosimeter under the simplest of circumstances for both coplanar and non-coplanar treatment deliveries. OSLD were irradiated in the spherical phantom, and the responses of the dosimeter from edge-on angles were normalized to the response when irradiated with the beam incident normally on the surface of the dosimeter. The average normalized response was used to establish an angular correction factor for 6 MV and 18 coplanar treatments, and for 6 MV non-coplanar treatments specific to CyberKnife. The RPC pelvic phantom dosimetry insert was modified to hold OSLD, in addition to the TLD, adjacent to the planes of film. Treatment plans of increasing angular beam delivery were developed, three in Pinnacle v9.0 (4-field box, IMRT, and VMAT) and one in Accuray’s MultiPlan v3.5.3 (CyberKnife). The plans were delivered to the pelvic phantom containing both TLD and OSLD in the target volume. The pelvic phantom was also sent to two institutions to be irradiated as trials, one delivering IMRT, and the other a CyberKnife treatment. For the IMRT deliveries and the two institution trials, the phantom also included film in the sagittal and coronal planes. The doses measured from the TLD and OSLD were calculated for each irradiation, and the angular correction factors established from the spherical phantom irradiations were applied to the OSLD dose. The ratio of the TLD dose to the angular corrected OSLD dose was calculated for each irradiation. The corrected OSLD dose was found to be within 1% of the TLD measured dose for all irradiations, with the exception of the in-house CyberKnife deliveries. The films were normalized to both TLD measured dose and the corrected OSLD dose. Dose profiles were obtained and gamma analysis was performed using a 7%/4 mm criteria, to compare the ability of the OSLD, when corrected for the angular dependence, to provide equivalent results to TLD. The results of this study indicate that the OSLD can effectively be used as a replacement for TLD in the RPC’s anthropomorphic QA phantoms for coplanar treatment deliveries when a correction is applied for the dosimeter’s angular dependence.

Determinación del coeficiente de nascencia en coberturas cespitosas : oasis del Norte de Mendoza (Argentina)

Uno de los principales problemas que se presenta para la investigación en césped acerca de la adaptabilidad de diferentes especies y sus mezclas o la introducción de otras nuevas, consiste en la dificultad de calcular la densidad de plantas, en forma lo suficientemente aproximada a los resultados reales como para permitir comparaciones estadísticamente válidas de la calidad o usos del cultivo resultante. Se propone una nueva metodología para reemplazar aquella tradicional de cálculo de necesidades de semilla por peso, con densidades determinadas por prueba y error según el resultado final por aspecto y corrección de cantidad de mezcla sólo por valor cultural. La propuesta considera cálculos matemáticos para determinar la cantidad exacta de semillas a emplear para obtener una densidad final de siembra controlada, teniendo en cuenta peso específico de la semilla y valor cultural, afectando el dato resultante por un factor de corrección (CN) que relaciona el valor teórico obtenido con la respuesta a campo. Es decir que la cantidad de semilla a emplear debe calcularse en función de la densidad de plantas deseada. En el presente trabajo el factor CN se determina para diferentes especies. Se parte de la hipótesis que es posible mejorar las formas de selección de especies y preparación de mezclas a través del cálculo de la necesidad de semillas por cm2 y que algunas de las especies utilizadas poseen baja eficiencia en la relación semilla-planta, por lo que se debería ajustar su proporción en las mezclas comerciales locales según este factor de cultivo. Como metodología se propone: a. análisis de poder germinativo, pureza y peso específico de semillas comerciales; b. con cuatro repeticiones, en bloques al azar, realizar siembras individuales de las especies en condiciones ideales de sustrato, iluminación y humedad, y siembra en las mismas condiciones de campo que un cultivo tradicional en condiciones de especie pura y consociadas (mezclas); c. conteo del número de plántulas obtenidas en cada caso por método propuesto por Lush y Franz; d. con los resultados, estimar los coeficientes (CN) promedio y sus respectivos intervalos de confianza. Esta nueva manera de calcular las cantidades exactas de cada componente de una mezcla para césped abre un campo muy importante a la investigación de especies más adecuadas a cada ambiente, dado que el método tradicional conduce a resultados muy ambiguos y de difícil comparación.

Análisis de la mortalidad en la unidad de cuidados intensivos del Hospital Central de Mendoza, Argentina

Objetivo: Analizar la mortalidad en la Unidad de Cuidados Intensivos (UCI) del Hospital Central de Mendoza y evaluar el valor predictivo de la escala APACHE II (Evaluación Fisiológica Aguda y de Salud Crónica). Material y Método: Se realizó un estudio retrospectivo y observacional de los pacientes ingresados a la Unidad de Cuidados Intensivos del Hospital Central de Mendoza, desde el 01/11/06 hasta el 31/03/08. Se calculó la distribución de sexos y de edades de la muestra, la estadía promedio, principales motivos de ingreso a la UCI y la puntuación APACHE II en las primeras 24 horas de internación. Se calculó la mortalidad esperada y la mortalidad obtenida global y se analizó el coeficiente entre ambas mortalidades. Resultados: Se incluyeron 904 pacientes, 61,82% masculinos y 38,18% femeninos, con una edad media 46 años (±19,36). Estadía promedio en la UCI 8,5 días promedio. El principal motivo de internación fueron los Traumatismos Encéfalocraneanos (TEC) con un 27,7% del total (86% asociados a politraumatismo grave). La mortalidad global obtenida fue del 41,48% vs. 24,08% esperable, con un coeficiente de mortalidad de 1,72 (p<0,0001). Conclusiones: La UCI estudiada presenta por las características de la población asistida un elevado índice de mortalidad global. La mortalidad obtenida fue 72% mayor a la mortalidad esperable según la puntuación APACHE II, demostrando esta Escala un bajo valor predictivo en nuestra UCI. La diferencia entre mortalidades podría parcialmente explicarse por la alta prevalencia de entidades con mortalidades subvaloradas por este modelo pronóstico, como pacientes politraumatizados y neurocríticos. En nuestro estudio, la Escala APACHE II presentó una franca subestimación de la mortalidad en ambas patologías. Sugerimos la realización de un estudio de regresión logística local para determinar un factor de corrección y/o adicionar puntos al valor APACHE II según el diagnóstico de ingreso del paciente. Asimismo, proponemos evaluar el empleo de medidas alternativas para predecir mortalidad, como sistemas de tercera generación (por ejemplo: APACHE III, MPM II y SAPS II).

Stable isotope record of planktonic foraminifera from the Sulu Sea

A high-resolution, accelerator mass spectroscopy 14C dated sediment record from the Sulu Sea clearly indicates that the Younger Dryas event affected the western equatorial Pacific. Planktonic foraminiferal delta18O and abundance data both record significant changes during Younger Dryas time. In particular, a 0.4 per mil increase in the delta18O value of Globigerinoides ruber and the reappearance of the cool water planktonic foraminifera, Neogloboquadrina pachyderma, occur during the Younger Dryas at this location. These isotopic and faunal changes are a response to either surface water temperature or salinity changes, or some combination of the two. Changes in surface salinities could have been accomplished through either local or global processes. Intensification of the monsoon climate system and increased precipitation at approximately 11 ka is one mechanism that may have resulted in local changes in salinity. A meltwater pulse derived from the Tibetan Plateau is another mechanism which may have caused local changes in salinity. The presence of the Younger Dryas in the tropical western Pacific clearly indicates that this climatic event is not restricted to the North Atlantic or high latitudes, but rather is global in extent.

Carbon/oxygen ratios, clay composition and gas-hydrate saturation of ODP Leg 164 sites

The analyses of downhole log data from Ocean Drilling Program (ODP) boreholes on the Blake Ridge at Sites 994, 995, and 997 indicate that the Schlumberger geochemical logging tool (GLT) may yield useful gas hydrate reservoir data. In neutron spectroscopy downhole logging, each element has a characteristic gamma ray that is emitted from a given neutron-element interaction. Specific elements can be identified by their characteristic gamma-ray signature, with the intensity of emission related to the atomic elemental concentration. By combining elemental yields from neutron spectroscopy logs, reservoir parameters including porosities, lithologies, formation fluid salinities, and hydrocarbon saturations (including gas hydrate) can be calculated. Carbon and oxygen elemental data from the GLT was used to determine gas hydrate saturations at all three sites (Sites 994, 995, and 997) drilled on the Blake Ridge during Leg 164. Detailed analyses of the carbon and oxygen content of various sediments and formation fluids were used to construct specialized carbon/oxygen ratio (COR) fan charts for a series of hypothetical gas hydrate accumulations. For more complex geologic systems, a modified version of the standard three-component COR hydrocarbon saturation equation was developed and used to calculate gas hydrate saturations on the Blake Ridge. The COR-calculated gas hydrate saturations (ranging from about 2% to 14% bulk volume gas hydrate) from the Blake Ridge compare favorably to the gas hydrate saturations derived from electrical resistivity log measurements.

Pliocene and Pleistocene eastern equatorial Pacific Mg/Ca subsurface temperature

During the early Pliocene warm period (~4.6-4.2 Ma) in the Eastern Equatorial Pacific upwelling region, sea surface temperatures were warm in comparison to modern conditions. Warm upwelling regions have global effects on the heat budget and atmospheric circulation, and are argued to have contributed to Pliocene warmth. Though warm upwelling regions could be explained by weak winds and/or a deep thermocline, the temporal and spatial evolution of the equatorial thermocline is poorly understood. Here we reconstruct temporal and spatial changes in subsurface temperature to monitor thermocline depth and show the thermocline was deeper during the early Pliocene warm period than it is today. We measured subsurface temperature records from Eastern Equatorial Pacific ODP transect Sites 848, 849, and 853 using Mg/Ca records from Globorotalia tumida, which has a depth habitat of ~50-100 m. In the early Pliocene, subsurface temperatures were ~4-5°C warmer than modern temperatures, indicating the thermocline was relatively deep. Subsurface temperatures steeply cooled ~2-3°C from 4.8 to 4.0 Ma and continued to cool an additional 2-3°C from 4.0 Ma to present. Compared to records from other regions, the data suggests the pronounced subsurface cooling between 4.8 and 4.0 Ma was a regional signal related to restriction of the Isthmus of Panama, while continued cooling from 4.0 Ma to present was likely related to global processes that changed global thermocline structure. Additionally, the spatial evolution of the equatorial thermocline along a N-S transect across ODP Sites 853, 849 and 848 suggests an intensification of the southeast trades from the Pliocene to present. Large-scale atmospheric and oceanographic circulation processes link high and low latitude climate through their influence on equatorial thermocline source water regions and consequently the equatorial thermocline. Through these low latitude/high latitude linkages, changes in the equatorial thermocline and thermocline source water played an important role in the transition from the warm Pliocene to the cold Pleistocene.

Reflectances transect from coast to open sea, Baltic Sea

In this study, retrievals of the medium resolution imaging spectrometer (MERIS) reflectances and water quality products using 4 different coastal processing algorithms freely available are assessed by comparison against sea-truthing data. The study is based on a pair-wise comparison using processor-dependent quality flags for the retrieval of valid common macro-pixels. This assessment is required in order to ensure the reliability of monitoring systems based on MERIS data, such as the Swedish coastal and lake monitoring system (http.vattenkvalitet.se). The results show that the pre-processing with the Improved Contrast between Ocean and Land (ICOL) processor, correcting for adjacency effects, improve the retrieval of spectral reflectance for all processors, Therefore, it is recommended that the ICOL processor should be applied when Baltic coastal waters are investigated. Chlorophyll was retrieved best using the FUB (Free University of Berlin) processing algorithm, although overestimations in the range 18-26.5%, dependent on the compared pairs, were obtained. At low chlorophyll concentrations (< 2.5 mg/m**3), random errors dominated in the retrievals with the MEGS (MERIS ground segment processor) processor. The lowest bias and random errors were obtained with MEGS for suspended particulate matter, for which overestimations in te range of 8-16% were found. Only the FUB retrieved CDOM (Coloured Dissolved Organic Matter) correlate with in situ values. However, a large systematic underestimation appears in the estimates that nevertheless may be corrected for by using a~local correction factor. The MEGS has the potential to be used as an operational processing algorithm for the Himmerfjärden bay and adjacent areas, but it requires further improvement of the atmospheric correction for the blue bands and better definition at relatively low chlorophyll concentrations in presence of high CDOM attenuation.