Systematic development and validation of a predictive model for major postoperative complications in the Peri-operative Quality Improvement Project (PQIP) dataset

Complications are common following major surgery and are associated with increased use of healthcare resources, disability and mortality. Continued reliance on mortality estimates risks harming patients and health systems, but existing tools for predicting complications are unwieldy and inaccurate. We aimed to systematically construct an accurate pre-operative model for predicting major postoperative complications; compare its performance against existing tools; and identify sources of inaccuracy in predictive models more generally. Complete patient records from the UK Peri-operative Quality Improvement Programme dataset were analysed. Major complications were defined as Clavien-Dindo grade ≥ 2 for novel models. In a 75% train:25% test split cohort, we developed a pipeline of increasingly complex models, prioritising pre-operative predictors using Least Absolute Shrinkage and Selection Operators (LASSO). We defined the best model in the training cohort by the lowest Akaike's information criterion, balancing accuracy and simplicity. Of the 24,983 included cases, 6389 (25.6%) patients developed major complications. Potentially modifiable risk factors (pain, reduced mobility and smoking) were retained. The best-performing model was highly complex, specifying individual hospital complication rates and 11 patient covariates. This novel model showed substantially superior performance over generic and specific prediction models and scores. We have developed a novel complications model with good internal accuracy, re-prioritised predictor variables and identified hospital-level variation as an important, but overlooked, source of inaccuracy in existing tools. The complexity of the best-performing model does, however, highlight the need for a step-change in clinical risk prediction to automate the delivery of informative risk estimates in clinical systems.

Development and internal validation of a model for postoperative morbidity in adults undergoing major elective colorectal surgery: the peri-operative quality improvement programme (PQIP) colorectal risk model

Over 1.5 million major surgical procedures take place in the UK NHS each year and approximately 25% of patients develop at least one complication. The most widely used risk-adjustment model for postoperative morbidity in the UK is the physiological and operative severity score for the enumeration of mortality and morbidity. However, this model was derived more than 30 years ago and now overestimates the risk of morbidity. In addition, contemporary definitions of some model predictors are markedly different compared with when the tool was developed. A second model used in clinical practice is the American College of Surgeons National Surgical Quality Improvement Programme risk model; this provides a risk estimate for a range of postoperative complications. This model, widely used in North America, is not open source and therefore cannot be applied to patient populations in other settings. Data from a prospective multicentre clinical dataset of 118 NHS hospitals (the peri-operative quality improvement programme) were used to develop a bespoke risk-adjustment model for postoperative morbidity. Patients aged ≥ 18 years who underwent colorectal surgery were eligible for inclusion. Postoperative morbidity was defined using the postoperative morbidity survey at postoperative day 7. Thirty-one candidate variables were considered for inclusion in the model. Death or morbidity occurred by postoperative day 7 in 3098 out of 11,646 patients (26.6%). Twelve variables were incorporated into the final model, including (among others): Rockwood clinical frailty scale; body mass index; and index of multiple deprivation quintile. The C-statistic was 0.672 (95%CI 0.660-0.684), with a bootstrap optimism corrected C-statistic of 0.666 at internal validation. The model demonstrated good calibration across the range of morbidity estimates with a mean slope gradient of predicted risk of 0.959 (95%CI 0.894-1.024) with an index-corrected intercept of -0.038 (95%CI -0.112-0.036) at internal validation. Our model provides parsimonious case-mix adjustment to quantify risk of morbidity on postoperative day 7 for a UK population of patients undergoing major colorectal surgery. Despite the C-statistic of < 0.7, our model outperformed existing risk-models in widespread use. We therefore recommend application in case-mix adjustment, where incorporation into a continuous monitoring tool such as the variable life adjusted display or exponentially-weighted moving average-chart could support high-level monitoring and quality improvement of risk-adjusted outcome at the population level.

Probing the seismic cycle timing with coseismic twisting of subduction margins

Assessing the timing of great megathrust earthquakes is together crucial for seismic hazard analysis and deemed impossible. Geodetic instrumentation of subduction zones has revealed unexpected deformation patterns at subduction segments adjacent to those that hosted recent mega-earthquakes: coastal sites move landward with faster velocities than before the earthquake. Here, we show observations from the largest and best-monitored megathrust earthquakes, and from a scaled analog model, to reveal that these events create coseismic and postseismic deformation patterns typical of a complete gear-like rotation about a vertical axis, hereafter called twisting. We find that such twisting alters the interseismic velocity field of adjacent subduction segments depending on the time since the last earthquake. Early interactions accelerate while late interactions decelerate local kinematics. This finding opens the possibility of using megathrust earthquakes, the characteristics of the twisting pattern, and the ensuing geodetic velocity changes, as a proxy for estimating the timing of the seismic cycle at unruptured segments along the margin.

Satellite geodesy and downscaled laboratory experiments reveal that great subduction earthquakes trigger step changes in kinematics of neighboring segments. This signal is potentially informative of the timing of the seismic cycle.

Assessment of the robustness of volumetric-modulated arc therapy for lung radiotherapy

Volumetric-modulated arc therapy (VMAT) is increasingly popular as a treatment method in radiotherapy owing to the speed with which treatments can be delivered. However, there has been little investigation into the effect of increased modulation in lung plans with regard to interfraction organ motion. This is most likely to occur where the planning target volume (PTV) lies within areas of low density. This paper aims to investigate the effect of modulation on the dose distribution using simulated patient movement and to propose a method that is less susceptible to such movement. Simulated interfraction motion is achieved by moving the plan isocentre in steps of 0.5 cm and 1.0 cm in six directions for five clinical VMAT patients. The proposed planning method involves optimisation using a density override of 1 g cm(-3), within the PTV in lung, to reduce segment boosting in the periphery of the PTV. This investigation shows that modulation can result in an increase in the maximum dose of >25%, an increase in PTV near-maximum dose of 17% and a reduction in near-minimum dose by 46%. Unacceptable organ at risk (OAR) doses are also seen. The proposed method reduces modulation, resulting in a maximum dose increase of 10%. Although safeguards are in place to prevent the increased dose to OARs from patient movement, there is nothing to prevent the increased dose as a result of modulation in lung. A simple planning method is proposed to safeguard against this effect. Investigation suggests that, where modulation exists in a plan, this method reduces it and is clinically viable.

Optimising stereotactic body radiotherapy for non-small cell lung cancer with volumetric intensity-modulated arc therapy--a planning study

AIMS

The potential advantages of stereotactic body radiotherapy (SBRT) for early stage non-small cell lung cancer (NSCLC) over conventional fractionated radiotherapy include a higher biological effective dose, a reduction in accelerated repopulation, greater patient convenience and reduced demand on radiotherapy resources. Before introducing SBRT in our department, a review of planning and delivery was undertaken, starting with an assessment of optimum beam number and arrangement.

MATERIALS AND METHODS

Radiotherapy planning computed tomography scans for five patients previously treated for T1 peripheral NSCLC were selected. In each the contoured tumour had planning target volume (PTV) margins of 1cm in all directions. Forward-planned three-field coplanar and non-coplanar plans and a seven-field coplanar plan were produced and optimised. In-house inverse-planning software (AutoBeam) was used to generate three-, five-, seven- and nine-field coplanar and non-coplanar plans and two volumetric intensity-modulated arc therapy (VMAT) plans. The resulting V(20), V(11), PTV(90), PTV(95) and mean lung dose were compared.

RESULTS

Analysis of variance showed non-coplanar plans to have lower V(11) and higher PTV(90) and PTV(95) than coplanar plans. VMAT showed equivalent V(20) and target coverage when compared with the best non-coplanar plans, but with a faster delivery time (2min 8s versus 12min 40s).

CONCLUSIONS

Inverse-planned five-field non-coplanar plans and VMAT improve target coverage while minimising the higher dose to normal lung tissue for SBRT of NSCLC compared with coplanar beam arrangements. VMAT is preferable because of significantly shorter treatment delivery times.

The effect of treatment position, prone or supine, on dose-volume histograms for pelvic radiotherapy in patients with rectal cancer

Patients undergoing radiotherapy for rectal cancer are generally treated in a prone position, with a full bladder, to reduce the volume of normal bowel in the high-dose volume. This position is difficult to maintain, and is not consistently reproducible. This study evaluates the volume of bowel and dose received in the prone and supine positions in patients undergoing pre-operative rectal cancer chemoradiation. Using CT planning, 19 consecutive patients with rectal cancer with a full bladder underwent CT scanning first in the prone position and then immediately afterwards in the supine position. The planning target volume was outlined for the prone position and transcribed to the supine scan using pre-set criteria. The bladder and small bowel were outlined in both positions. Radiotherapy was planned using three-dimensional conformal planning, and treatment was delivered using three fields with multileaf collimators in two phases: phase I, pelvis 45 Gy/25 fractions; and phase II, tumour 9 Gy/five fractions. For both positions, the volume of bowel receiving doses in 5 Gy increments from 5-45 Gy was calculated using dose-volume histograms. At 5 Gy and 10 Gy dose levels, a significantly higher volume of bowel was irradiated in the supine position (p<0.001). At 15 Gy, it was marginally significant (p = 0.018). From 20-45 Gy, there was no significant difference in the volume of bowel irradiated with each 5 Gy increment. This study demonstrates that the volume of bowel irradiated at doses associated with bowel toxicity in concurrent chemoradiation is not significantly higher in the supine position. This position could be adopted for patients undergoing pre-operative rectal cancer chemoradiation.

The use of gel dosimetry for verification of electron and photon treatment plans in carcinoma of the scalp

In recent years there has been a large amount of research into the potential use of radiation sensitive gels for three-dimensional verification of clinical radiotherapy doses. In this paper we report the use of a MAGIC gel dosimeter (Fong et al 2001 Phys. Med. Biol. 46 3105) for the verification of a specific patient's radiation therapy dose distribution. A 69-year-old male patient presented with a squamous cell carcinoma extending approximately 180 degrees across the top of the scalp (anterior to posterior) and from just over midline to 90 degrees left of the skull. The patient's treatment was commenced using two electron fields. For gel dosimetry, phantoms were produced in which the outer surface spatially corresponded to the outer contours of the patient's anatomy in the region of irradiation. The phantoms were treated with either electrons or intensity modulated radiation therapy (IMRT) with photons. The results identified a hot spot between the matched electron fields and confirmed the more homogeneous dose distribution produced by the IMRT planning system. The IMRT plan was then clinically implemented. The application of a clinical dose to a phantom shaped to a specific patient as well as the ability to select a slice at will during phantom imaging means that gel dosimetry can no longer be considered to simply have potential alone, but is now in fact a useful dosimetric tool.

Eating attitudes across age and gender groups: a Canadian study

OBJECTIVE

The purpose was to examine the eating attitudes across gender and age groups.

METHOD

The study included 343 individuals over 18 years of age (range: 18 to 94 years; mean 48 +/- 22), who were divided into 3 age categories: 18 to 34 years, 35 to 64 years and 65 years and above with a distribution of 35% (119), 37% (128) and 28% (97) respectively. The participants completed a background questionnaire and the Eating Attitudes Test (EAT-26).

RESULTS

A total of 30 participants (8.7%) scored above the cut-off point on the EAT-26 which is indicative of potential eating disturbances. Only statistically significant gender differences were observed in the total EAT-26 scores. Factor analysis revealed three subscales of EAT-26, food preoccupation, image preoccupation, and external focus or social pressure, with acceptable level of reliability and these were used for age and gender comparison. Age and gender differences were observed in the food and image preoccupation subscales only.

DISCUSSION

Although EAT-26 is widely used in clinical and non-clinical population, age and gender comparisons have not been made in a population-based sample. Use of the subscales of eating attitudes should help health professionals better identify those at risk for eating disturbances and provide the opportunity for intervention.

A versatile image analysis approach for simultaneous chromosome identification and localization of FISH probes

Modern cytogenetic techniques, such as comparative genomic hybridization (CGH) and the multi-color fluorescence in situ hybridization (FISH) techniques of multiplex fluorescence in situ hybridization (M-FISH) and spectral karyotyping (SKY), require a coordinated banding analysis to maximize their usefulness. All of the methods currently used, including Giemsa (G-) banding, Alu banding, and 4',6-diamidino-2-phenyl-indole (DAPI) banding, have serious drawbacks. A simple and effective method to band chromosomes concurrently with FISH is needed. To address this problem, we stained chromosomes with DAPI and chromomycin A3, and then used an image analysis program to generate banding by dividing the image taken with a DAPI excitation filter by the image taken with a chromomycin A3 excitation filter. The result was a metaphase spread in which the chromosomes possessed a banding pattern characteristic of R-banding. The image analysis program was then used to generate linescans of pixel intensity versus relative position along the length of chromosomes that were banded using this technique, which we have called D/C R-banding. Each chromosome in a genome was represented by a characteristic scan profile, which was unaffected by FISH signals. Reference linescans were prepared by karyotyping D/C R-banded chromosomes for a given species, and then drawing lines along the length of the known chromosomes. The linescans were combined into a spreadsheet database, which was linked by dynamic data exchange to the image analysis program and normalized for length and intensity. The linescan of an unknown chromosome was then transferred to the spreadsheet, where it was normalized for length and intensity and overlaid on the linescans of each chromosome in the genome. Unknown chromosomes were identified by comparison of their graphs with graphs in the standardized reference genome. We have used this approach to create reference linescan karyotypes of several species, and to identify chromosomes on which FISH was performed.

Influenza vaccination in asthma: a primary care experience

BACKGROUND

Despite the recommendation of the Department of Health that patients with asthma receive annual vaccination against influenza, uptake remains unsatisfactory with many patients suspicious that vaccination is harmful.

AIM

To examine the effect of influenza vaccination on asthmatic patients typical of a general practice setting.

METHOD

A multicentre study with 56 patients participating from 14 practices in England and Scotland. Patients completed peak expiratory flow rate (PEFR) and symptom diaries for two weeks before and two weeks after influenza vaccination.

RESULTS

A non-significant fall in baseline PEFR of 10.5 l/min, from an average of 431.5 l/min, was observed after influenza vaccination, representing a 2% change from baseline. A significant increase in night time reliever use of 0.17 puffs per night (P < 0.01) was found. Non-significant increases in number of nights per week with sleep disturbed due to asthma, severity of night-time and day-time symptoms, and day-time reliever use were also noted.

CONCLUSION

Influenza is an important cause of morbidity and mortality in asthmatics. This study confirms the safety of influenza vaccination in patients with asthma typical of those seen in primary care. General practitioners need not hesitate in recommending this valuable intervention to their asthmatic patients and should consider ways in which uptake can be improved.

Topographic mapping of the pattern onset evoked magnetic response to stimulation of different portions of the visual field

Visual evoked magnetic responses to pattern onset/offset stimuli were recorded in 5 normal subjects. The outgoing and ingoing magnetic currents were seen maximally over the contralateral visual cortex to the half field stimulated in the majority of subjects. The distribution of magnetic currents with half field stimulation would be consistent with a dipole on the midline with the positive end pointing laterally. Quadrantic stimuli produced dipoles which would be consistent with a vector sum of dipoles situated, in the case of upper quarter field stimulation, in the lingual gyri on the floor of the calcarine fissure. Lower quadrantic stimulation produced a vector consistent with dipoles on the cuneal gyrus and on the ceiling of the calcarine fissure producing a resultant vector. The use of paired octant stimuli to improve localisation was demonstrated on both right left and upper and lower half fields.

Putative determinants of the cellular response to hyperthermia

Recently, it has been demonstrated that two different thermal resistant states found in Chinese hamster cells, one transient, associated with thermotolerance, and the other permanent, associated with the increased expression of the cognate member of the hsp 70 family, are characterized by faster recovery from heat-induced perturbations in several cellular processes (Laszlo 1992b). These processes include total cellular protein and RNA synthesis, the localization of hsp70, the organization of vimentin, and the protein composition of the nucleus. In the present study, the recovery from heat-induced perturbations in cellular physiology was extended further to two more types of Chinese hamster cells: permanently heat resistant cells in which thermoresistance is associated with the overexpression of hsp27 and heat-sensitive cell lines. When the heat-resistant hsp27 transfected cell lines were compared with the control wild-type cell line, the recovery of protein synthesis from heat-induced inhibition was similar in the normal and hsp27 transfected cells, while the recovery from heat-induced inhibition of total RNA synthesis and the recovery from heat-induced increased association of hsp70 with nuclei were both more rapid in the hsp27 transfected cell lines. In the permanently heat-sensitive cell lines, the kinetics of recovery from heat-induced inhibition of protein synthesis did not correlate with the heat sensitive state. However, delays in the recovery from heat-induced alterations in total cellular RNA synthesis and from heat-induced excess nuclear association of hsp70 were associated with the heat-sensitive state. Overall, these results suggest that the kinetics of recovery from heat-induced alterations in total cellular RNA synthesis and the localization of hsp 70 are putative candidates for being determinants of the cellular response to hyperthermia, and thus have the potential to form the basis of predictive assays for use in conjunction with clinical hyperthermia.

Amino acid structure and "difficult sequences" in solid phase peptide synthesis

Deprotection peak profiles have been determined as a measure of internal aggregation during Fmoc-polyamide continuous flow solid phase synthesis. The results have been correlated with amino-acid structure and discussed in terms of minimising aggregation during synthesis.

A sensitive radioimmunoassay measuring endothelin-like immunoreactivity in human plasma: comparison of levels in patients with essential hypertension and normotensive control subjects

1. A radioimmunoassay has been developed for measuring endothelin-like immunoreactivity in human plasma using an antibody raised against endothelin-1 which also cross-reacts with big endothelin-1 and endothelin-2 but not endothelin-3. 2. The sensitivity of the assay was 1 fmol/tube with inter- and intra-assay coefficients of variation of 13% and 9%, respectively. Cross-reactivity with endothelin-3 and non-endothelin peptides was less than 1%. 3. Endothelin-like immunoreactivity was present in the plasma of hypertensive patients (n = 25) at a concentration of 5.7 +/- 0.5 pmol/l (mean +/- SEM), which was not significantly different from that of age-matched control subjects (5.1 +/- 0.5 pmol/l). At these levels, endothelin-1 is unlikely to function as a circulating hormone. 4. Within the normotensive group, the concentration of endothelin-like immunoreactivity in plasma was positively correlated with mean arterial blood pressure, but in hypertensive patients it showed a significant negative correlation.

Basic radiobiology

Experimental studies of the biological effects of radiation were started soon after the discoveries of x-rays in 1895, but there is still much that is not known. This article includes some research objectives that are essentially pragmatic in nature, intended to support and improve the current practice of radiotherapy, but the central thrust is the understanding of the mechanisms involved in the biological effects of radiation at the cellular and molecular levels. The article was written by a consortium of scientists and suffers inevitably from the drawback that writing styles are inconsistent, and coverage is not uniform. However, it benefits from the enormous advantage that it reflects the accumulated wisdom and judgment of more than a dozen scientists who, in their own areas of expertise, are recognized as being at the cutting edge of radiation research. The niceties of style and syntax are sacrificed in favor of the quality of the science and the maturity of judgment. The study of DNA damage as a mechanism for cell injury in early- and late-responding tissues, as well as a comparison of DNA damage that leads to lethality, as opposed to transformation and mutagenesis, are key items. The study of cell lethality with cells in culture led to the identification of repair, both sublethal and potentially lethal, as well as the dose-rate effect, and has had a considerable impact on radiotherapy. Future studies should focus on understanding the factors that determine radiosensitivity/radioresistance. A variety of approaches are available, including the study of genetically deficient cell lines from cancer-prone individuals. A parallel approach is the application of the techniques of molecular biology to clone the repair genes in mammalian cells, and to understand genetic defects that alter gene regulation, or to regulate biochemical factors in the cell. Substantial progress has been made in developing in vitro assays for mutagenesis, particularly using hybrids of rodent and human cells. Better methods are needed to study the effects of mutation on gene expression, and sensitive systems are needed that can detect low doses of radiation. Assays of oncogenic transformation, the in vitro counterpart of carcinogenesis, have been used to investigate the oncogenic potential of various types of radiation and chemotherapy agents. Key topics in future will include the investigation of supra-additivity between different agents, the identification and characterization of oncogenes that may be activated by radiation, the development of quantitative assays based on human cells, and further studies involving cell-to-cell communication.(ABSTRACT TRUNCATED AT 250 WORDS)

Duplication of part of the long arm of chromosome 1 in marrow cells of a treated case of myelomatosis

In a case of classical myelomatosis treated with melphalan, a clone of cells with a chromosomal abnormality was found in the bone marrow during remission. There was good reason to think that the hemopoietic cells, rather than plasmacytoma cells, were implicated. Although the clone persisted, no evidence of leukemia developed over a period of observation of 2 yr. The anomaly was interpreted as a duplication of part of the long arm of chromosome 1, which appeared to involve the segment q21 to q31.

A patient with 45,X-46,XXq--46,XXq-dic karyotype

Images

Chromosomal studies in patients taking phenylbutazone

Images

t(2q-; Dq+) in a mentally retarded female child

Images