Auckland Hospital Research Articles

Effects of Neonatal Hyperglycemia on Retinopathy of Prematurity and Visual Outcomes at 7 Years of Age: A Matched Cohort Study

To determine whether neonatal hyperglycemia is associated with retinopathy of prematurity (ROP), visual outcomes, and ocular growth at 7years of age. Children born preterm (<30weeks of gestational age) at a tertiary hospital in Auckland, New Zealand, who developed neonatal hyperglycemia (2 blood glucose concentrations ≥153mg/dL [8.5mmol/L] 4hours apart) were matched with children who were not hyperglycemic (matching criteria: sex, gestational age, birth weight, age, socioeconomic status, and multiple birth) and assessed at 7years of corrected age. The primary outcome, favorable overall visual outcome (visual acuity ≤0.3 logarithm of the minimum angle of resolution, no strabismus, stereoacuity ≤240 arcsec, not requiring spectacles) was compared between groups using generalized matching criteria-adjusted linear regression models. Assessments were performed on 57 children with neonatal hyperglycemia (hyperglycemia group) and 54 matched children without hyperglycemia (control group). There were no differences in overall favorable visual outcome (OR 0.95, 95% CI 0.42-2.13, P=.90) or severe ROP incidence (OR 2.20, 95% CI 0.63-7.63, P=.21) between groups. Children with hyperglycemia had poorer binocular distance visual acuity (mean difference 0.08, 95% CI 0.03-0.14 logarithm of the minimum angle of resolution, P<.01), more strabismus (OR 6.22, 95% CI 1.31-29.45, P=.02), and thicker crystalline lens (mean difference 0.14, 95% CI 0.04-0.24mm, P<.01). Maximum blood glucose concentration was greater in the ROP-treated group compared with the ROP-not treated and no ROP groups after adjusting for sex, gestational age, and birth weight z score (P=.02). Neonatal hyperglycemia was not associated with overall visual outcomes at 7years of age. However, there were between-group differences for specific outcome measures relating to interocular lens growth and binocular vision. Further follow-up is required to determine implications on long-term visual outcome.

Primary medication non-adherence at Counties Manukau Health Emergency Department (CMH-ED), New Zealand: an observational study

ObjectivesTo measure primary medication non-adherence (failure to fill prescription medicines) in patients discharged from the emergency department (ED), and to determine whether sociodemographic factors, smoking status and access to a...

The “unfortunate experiment” that was not, and the indebtedness of women and children to Herbert (“Herb”) Green (1916–2001)

All screening programs have the potential to result in harmful overdiagnosis and overtreatment. Sound evidence, policy, and standards are needed to keep this harm to a minimum. Screening for and “treating” cervical cytological abnormalities provides a sobering example. The term “carcinoma in situ” came into use from 1950 and implied—misleadingly—that an inevitable malignant process had been identified. Hysterectomy became widely used to “treat” it. New Zealand's Herbert Green was one of a minority of gynecologists around the world who recognized the possible dangers of harmful overtreatment. Green developed and monitored more conservative management approaches to avoid women being “subjected to hysterectomy”. By the mid-1980s, his approaches had been adopted not only within National Women's Hospital in Auckland, but more widely.In 1987, it was alleged in a magazine article that an “experimental” research program had been undertaken at National Women's Hospital to study the natural course of untreated cervical “carcinoma in situ”; that this had entailed withholding “conventional treatment” from some patients indefinitely; and that some patients had died as a result. A public furore resulted and led 2 weeks later to the creation of a judicial inquiry—the Cartwright Inquiry—which reported the following year.The findings of the Inquiry, which criticized Green's practice, have been both disputed and defended ever since, often by individuals with competing interests. In the recently published 2nd edition of their book Screening: Evidence and Practice, Angela Raffle, Anne Mackie, and Muir Gray have provided a fresh analysis. This concludes (see accompanying article) that the Cartwright Inquiry provides no trustworthy evidence of harm from the adoption of the conservative management introduced by Green and adopted by some of his colleagues at National Women's Hospital. It is now clear that the treatment and monitoring methods introduced by Herb Green have benefited numerous women through avoidance of major surgery and preservation of fertility.

Human behaviour during a healthcare facility evacuation drills: Investigation of pre-evacuation and travel phases

Evacuation models are key tools to assess the fire safety of complex buildings. Those tools and their results rely on the input values selected by users based on existing datasets and the way they are transformed into output. Several evacuation studies have been carried out to provide input values for evacuation models. However, those regarding healthcare facilities evacuation are still rare in the literature.In this paper, we present a new evacuation dataset for healthcare facility evacuation simulation. The data was collected from an outpatients’ area of a public hospital in Auckland (New Zealand) during two unannounced fire drills. The video images were analysed to generate new evacuation model inputs for healthcare facility evacuation scenarios. The drills involved both staff and patients. Pre-evacuation times, evacuee horizontal travel speeds, exit selection and total evacuation times were collected and analysed. Moreover, we investigated evacuee reactions and actions to study staff and patients’ interaction during the evacuation process.The results showed that pre-evacuation time of patients ranges from 8 to 63 s; while, pre-evacuation time of staff ranges from 8 to 141 s. In addition, during the movement phase, staff who were not assisting patients, and patients with no impairments, travelled at a similar average walking speed (i.e. 1.06 m/s for staff members and 0.93 m/s for patients with no impairments). Finally, the results indicated that the average travel speed of patients with walking impairments and staff assisting them was almost half of the travel speed of the first two groups (i.e. 0.52 m/s).

Evacuation Data from a Hospital Outpatient Drill The Case Study of North Shore Hospital

Assessing the fire safety of buildings is fundamental to reduce the impact of this threat on their occupants. Such an assessment can be done by combining existing models and existing knowledge on how occupants behave during fires. Although many studies have been carried out for several types of built environment, only few of those investigate healthcare facilities and hospitals. In this study, we present a new behavioural data-set for hospital evacuations. The data was collected from the North Shore Hospital in Auckland (NZ) during an unannounced drill carried out in May 2017. This drill was recorded using CCTV and those videos are analysed to generate new evacuation model inputs for hospital scenarios. We collected pre-movement times, exit choices and total evacuation times for each evacuee. Moreover, we estimated pre-movement time distributions for both staff members and patients. Finally, we qualitatively investigated the evacuee actions of patients and staff members to study their interaction during the drill. The results show that participants were often independent from staff actions with a majority able to make their own decision.

E-referrals and teledermatoscopy grading for melanoma: a successful model of care.

An e-referral system was developed at a tertiary care hospital in Auckland, New Zealand in 2014 for suspected cutaneous malignancy. E-referrals include patient information, a description of the lesion(s), biopsy results and/or attached photograph(s). Experienced surgical oncologists prioritised the referrals and selected a management option or referred them for a teledermatoscopy opinion. Our aim was to review the efficacy of e-referrals for improving diagnostic accuracy for melanoma. Referrals received in 2016 including images and categorisation as confirmed, likely or suspected melanoma by the triage specialist were evaluated. Concordance of the pathological diagnosis with the triage diagnosis and teledermatoscopy diagnosis was determined for each referral. 809 of 3470 e-referrals for skin cancer were categorised as confirmed, likely or suspected melanoma. 230 (28.4%) of these included a referral histopathology confirming melanoma/melanoma in situ. Of the remaining 579 referrals, 315 were sent for urgent diagnostic excision and 264 were referred for teledermatoscopy. 120 of the 315 sent for urgent excision were confirmed as melanoma (53) or melanoma in situ (67) on histopathology: a positive predictive value (PPV) of 38.1% and number needed to excise (NNE) of 2.6. Less than 10% of referrals triaged for teledermatoscopy were confirmed as melanoma (24/264). Almost half of all referrals (374/809, 45.6%) included melanoma/melanoma in situ. The melanoma: melanoma in situ ratio was 1: 1.18. The e-referral and teledermatoscopy service for suspected melanoma has proven fewer unnecessary excisions of benign lesions than previously reported.

Comparison of bowel dysfunction between colorectal cancer survivors and a non-operative non-cancer control group.

Low anterior resection syndrome (LARS) detrimentally affects quality of life in colorectal cancer survivors. This study assessed the prevalence for LARS in colorectal cancer survivors and the same symptoms in a matched control group. Validated instruments, the LARS score and Short Form Survey 12, used to collect functional and quality of life outcomes from patients who had undergone distal colorectal resection at Auckland Hospital (2008-2015) or Dunedin Hospital (2008-2017). A matched non-operative control group was drawn from patients undergoing surveillance colonoscopy. The response rate was 79%. Cross-sectional prevalence of major LARS in rectal cancer patients was 52% at a median follow-up of 52months. Major LARS prevalence in the sigmoid cancer resection and non-cancer control groups was similar (25% vs 26%, P=0.6). On univariate analysis anastomotic height [risk ratio (RR) for low anterior resection4.6, P<0.001; ultralow anterior resection RR=15.5, P<0.001], radiotherapy (RR=2.6; P=0.009), stoma (RR=3.6; P=0.001) and J pouch reconstruction (vs straight anastomosis, RR=4.6; P=0.008) were associated with major LARS for rectal cancer patients. These factors were not significant when the analysis was stratified for anastomotic height. Despite correlation between LARS and Short Form Survey 12 outcomes (physical ρ=-0.2; mental ρ=-0.2) there was no difference in quality of life outcomes between the groups. Bowel dysfunction after low anterior resection affects the majority of rectal cancer patients. The high background rate of bowel dysfunction must be considered when assessing the prevalence of LARS.

782 Acute Coronary Syndrome in Severe Chronic Kidney Disease – Analysis of Management and Outcome

Choice of medical therapy alone or invasive revascularisation in patients with severe chronic kidney disease (CKD) presenting with acute coronary syndrome (ACS) is difficult. Clinical trials frequently exclude these patients making optimal management controversial. We assess the management and outcome in Auckland, New Zealand. Patients with advanced CKD (stage 4 or 5) or end stage kidney disease (ESKD) on dialysis who presented to hospital in Auckland between 2012-2019 with ACS were identified from the national ANZACS-QI registry. Clinical details, management and combined endpoint of major adverse cardiovascular events (MACE) of all-cause mortality, stroke, rehospitalisation and heart failure were analysed with a median follow-up of 2.2±1.8 years. 229 patients identified (17% peritoneal, 26% haemodialysis). Mean age 65±11y. Male 80(34.9%). Polynesians 126(55%), Europeans 51(22.3%). STEMI 18% and NSTEACS 82%. Diabetic nephropathy predominates (60%). Over half has prior history of ACS and presents in heart failure. Those on dialysis were more likely to undergo invasive coronary angiography (ICA) (83% vs 65%, p=0.002). Overall rate of ICA is 73%, similar in both STEMI and NSTEACS. Stent revascularisation comprise 30% and coronary artery bypass grafting 17%. Peri-angiographic complications were low (significant bleed 1%, stroke 1%, contrast nephropathy 3.6%, death 1.8%). Patients who underwent ICA had reduced all-cause-mortality (OR 0.55, 95% CI 0.38-0.79, p=0.001) and MACE (OR 0.61, 95% CI 0.43-0.84, p=0.003) compared to medical therapy alone. In the Auckland region, over half of patients with advanced CKD presenting with ACS undergo ICA which appears to reduce mortality and MACE rates compared to medical therapy alone.

Body composition of New Zealand-born term babies differs by ethnicity, gestational age and sex

BackgroundBody composition provides important information on nutrition and future metabolic risk. New Zealand has a diverse ethnic population for which there are no newborn body composition data. AimTo determine body composition in a cohort of New Zealand-born term babies. Study designObservational study. SubjectsHealthy, term infants between 37+0 and 41+6 weeks' gestation in two hospitals in Auckland, New Zealand. Outcome measuresBody composition by air displacement plethysmography and anthropometry measured within 5 days of birth. Parent-identified ethnicity was prioritised according to Ministry of Health criteria. Data were analysed using t-test, ANOVA with Tukey post-hoc tests, quantile regression and are mean(SD). Results440 babies (54% male) were included. Pacific Island/Māori (PI/M) were heavier at birth than Asian/Middle Eastern/Latin American/African (Asian+) babies (3403(506) vs 3181(485) g, p < .05). PI/M and European (E) babies were longer with larger head and waist circumferences than Asian+ babies (all p < .05). Absolute fat mass (FM) was not different amongst ethnicities (E, 365(156), PI/M, 347(183), Asian+, 357(188) g) but PI/M babies had significantly lower FM% than Asian+ (9.8(4.3) vs 10.9(4.5) %, p < .05). Fat-free mass (FFM) was greater in PI/M (3056(400) g) than E (2952(345) g (p < .05) and both PI/M and E had greater FFM than Asian+ (2824(363) g, p < .05). Early term babies had less FFM than term and late-term babies (2732(370), 3012(352), 3173(302)g, p < .001) respectively. ConclusionsAsian+ babies were the smallest babies with the least FFM yet had similar FM and the highest FM%, indicative of a thin, fat phenotype from birth.

A Prospective Study of Cosmetic Outcomes for Patients Treated with Breast Conservation and Radiation Therapy using Shorter Fractionation Schedules at Auckland Hospital, New Zealand

AimsBreast cosmesis is an important outcome for women after breast conservation. It has been shown to be correlated with better patient satisfaction, sexuality and self-esteem. This study reports the subjective and objective breast cosmesis outcomes for women treated at Auckland Hospital Radiation Oncology Department using hypofractionated radiation therapy. Materials and methodsBreast cosmesis was evaluated using a subjective questionnaire and standardised photographs before radiation, at 6 weeks, 1 year, 3 years and 5 years after radiation. Objective evaluation of the photographs and completion of questionnaires were undertaken by clinicians at the same time points. The questionnaire evaluated global cosmesis on a four-point scale – excellent (E), good (G), fair (F) and poor (P). Patient, tumour and treatment factors were evaluated to assess the impact on cosmesis. ResultsThree hundred and fifty-eight patients completed the baseline assessments and two hundred and two patients completed the 5-year assessments. Sixty-eight per cent of patients at baseline and 70% at 5 years scored their cosmesis as E/G compared with 52% and 51%, respectively, by clinician assessment. Age >50 years, separation ≤25 cm, non-diabetic, T1 tumours, node negative, quadrant of the scar, no boost and no adjuvant endocrine therapy were correlated with E/G cosmesis. On multivariate analysis at baseline, the individual breast factors that were significantly associated with a score of E/G cosmesis were breast shape (P = 0.028) and scar appearance (P = 0.001). At 5 years, breast shape (P = 0.003), nipple shape (P = 0.019) and scar appearance (P = 0.001) were found to be significant. ConclusionThis study shows that most women reported no significant change in their breast cosmesis after hypofractionated radiation treatment over 5 years and that a number of patient, tumour and treatment factors may impact on breast cosmesis.

234P - Implementation of mainstream BRCA testing in epithelial ovarian cancer in a tertiary centre

BackgroundThe literature reports that a minority of women with epithelial ovarian cancer (EOC) are referred for germline BRCA testing despite implications for treatment and cancer prevention. Mainstream testing is BRCA testing undertaken by the Medical Oncology service, rather than clinical genetics; and it was introduced at Auckland Hospital in 2017. This study assessed whether the implementation of mainstreaming increased the uptake and timeliness of BRCA testing in a multi-cultural tertiary centre. MethodsA retrospective analysis of women with high grade non-mucinous EOC attending Medical Oncology clinic was performed prior to and after the introduction of mainstream testing. Eligibility for BRCA testing included those with the above diagnosis ≤ 70 years (tested by Medical Oncology after mainstreaming introduced) and patients > 70 years with a personal and/or relevant family history (tested by genetics). The primary outcome was the proportion of eligible patients who underwent BRCA testing. A secondary outcome was the time to the availability of the BRCA test result. ResultsThe proportion of eligible women tested for germline BRCA mutations significantly increased (Table). The improvement was due to increased testing via mainstreaming; there was no change in the proportion of women tested through the Genetics service. Efficiency of testing also improved; the median time from Genetics referral to an available BRCA result was 146 days (range 37 to 1599 days) and from Medical Oncology assessment to BRCA test result 64 days (range 34 to 471 days) before and after mainstreaming respectively. Overall, 14.9% of women had germline BRCA 1 or 2 mutations.Table234PUnlabelled ImagePre- mainstreamingPost- mainstreamingP valueEligible for mainstreaming*108/177 (61.0%)37/40 (92.5%)0.0001>70 years + personal/ family history**10/15 (66.7%)7/9 (77.8%)0.6687Overall118/192 (61.5%)44/49 (89.8%)0.0001*Pre-mainstreaming, patients tested by Genetics; post-mainstreaming, tested in Medical Oncology clinic.**Tested by Genetics. ConclusionsMainstream testing for germline BRCA mutations improved quality – access and timeliness – for women with EOC. Efficiency of testing will have important therapeutic implications. Legal entity responsible for the studyAuckland District Health Board. FundingAuckland District Health Board. DisclosureAll authors have declared no conflicts of interest.

Regional healthcare costs and burden of injury associated with electric scooters

BackgroundThe introduction of shared electric scooters (e-scooters) to New Zealand has resulted in a large number of injuries. Within the past year, there have been studies addressing some of the impact of these e-scooter injuries, but none have included outpatient data or total regional costs. MethodsThis was a retrospective review of e-scooter associated injuries presenting to Auckland region healthcare providers in the seven months since their introduction using Accident Compensation Corporation (ACC) Claims data. The type of injuries and key metrics of their overall hospital burden were assessed between September 2018 and April 2019. The financial cost of these injuries was also estimated. ResultsA total of 770 patient presentations associated with e-scooters were identified during the study period. Of these, 524 (68.1%) were treated in the community by primary care physicians and 246 (31.9%) were treated in Auckland hospitals. The 246 hospital presentations used a total of 5,569 hospital bed-hours with 75 patients (30.5%) requiring admission and inpatient care. Of the hospital presentations, 49 patients (19.9%) required at least one operation, and 105 (42.7%) required specialist follow up care. 26.8% of injuries were thought to be associated with alcohol use. The estimated injury rate was 60 per 100,000 trips and hospital presentation rate was 20 per 100,000 trips. The combined cost attributable to these injuries was $608,843 (NZD) for Auckland City Hospital and $1,303,155 for the whole Auckland region. ConclusionsThe overall burden of care due to the introduction of e-scooters to New Zealand has had significant impact both on the primary urban trauma center as well as community care facilities. E-scooter related injuries have had a large impact on regional healthcare costs.

The multi-dimensional matrix for consultation-liaison psychiatry (mMAX-LP).

Consultation-liaison psychiatry (CLP) services are particularly susceptible to heterogeneity, developing haphazardly in response to local interests and perceived need. This hampers the generalisability of comparisons between services in terms of service models, resource requirements and outcome data. The objective of this paper therefore is to chronicle the development of a method to meaningfully describe, map and compare different CLP services. A review of the literature was followed by multiple site visits in both New Zealand and England, and an extended process of consultation and feedback. Sixteen dimensions common to CLP services were extracted to create a multi-dimensional matrix (mMAX-LP) which had three broad clusters (structure, coverage and relationship with physical health services). The model was applied and discussed with the previously visited hospitals over the succeeding five years. Additionally, the matrix was tested, and its utility demonstrated during the planned reconfiguration of CLP services at a large teaching hospital in South Auckland, New Zealand by tracking the evolution of CLP services. mMAX-LP shows promise as a useful model for profiling and comparing CLP services; mapping their evolution over time; and sign-posting future service development.

Status epilepticus in Auckland, New Zealand: Incidence, etiology, and outcomes.

To determine the incidence, etiology, and outcome of status epilepticus (SE) in Auckland, New Zealand, using the latest International League Against Epilepsy (ILAE) SE semiological classification. We prospectively identified patients presenting to the public or major private hospitals in Auckland (population = 1.61 million) between April 6, 2015 and April 5, 2016 with a seizure lasting 10minutes or longer, with retrospective review to confirm completeness of data capture. Information was recorded in the EpiNet database. A total of 477 episodes of SE occurred in 367 patients. Fifty-one percent of patients were aged <15 years. SE with prominent motor symptoms comprised 81% of episodes (387/477). Eighty-four episodes (18%) were nonconvulsive SE. Four hundred fifty episodes occurred in 345 patients who were resident in Auckland. The age-adjusted incidence of 10-minute SE episodes and patients was 29.25 (95% confidence interval [CI] = 27.34-31.27) and 22.22 (95% CI = 20.57-23.99)/100000/year, respectively. SE lasted 30minutes or longer in 250 (56%) episodes; age-adjusted incidence was 15.95 (95% CI = 14.56-17.45) SE episodes/100000/year and 12.92 (95% CI = 11.67-14.27) patients/100000/year. Age-adjusted incidence (10-minute SE) was 25.54 (95% CI = 23.06-28.24) patients/100 000/year for males and 19.07 (95% CI = 16.91-21.46) patients/100000/year for females. The age-adjusted incidence of 10-minute SE was higher in Māori (29.31 [95% CI = 23.52-37.14]/100000/year) and Pacific Islanders (26.55 [95% CI = 22.05-31.99]/100000/year) than in patients of European (19.13 [95% CI = 17.09-21.37]/100000/year) or Asian/other descent (17.76 [95% CI = 14.73-21.38]/100000/year). Seventeen of 367 patients in the study died within 30days of the episode of SE; 30-day mortality was 4.6%. In this population-based study, incidence and mortality of SE in Auckland lie in the lower range when compared to North America and Europe. For pragmatic reasons, we only included convulsive SE if episodes lasted 10minutes or longer, although the 2015 ILAE SE classification was otherwise practical and easy to use.

SAT-074 LIVE KIDNEY DONATION AOTEAROA - IMPROVING LIVE KIDNEY DONATION IN INDIGENOUS AND MIGRANT POPULATIONS IN SOUTH AUCKLAND, NEW ZEALAND

The Live Kidney Donation Aotearoa Project (LKDA) was a 4 year initiative to increase the number of people donating kidneys to patients with end stage kidney disease either before commencing dialysis or receiving dialysis treatment at a large metropolitan hospital in South Auckland, New Zealand. The project was a joint venture with a patient support group. The focus was on indigenous Maaori and migrant Pasifika people (PI), both donating and receiving kidneys, as there is a demonstrated disparity compared to the rest of the dialysis population that is not entirely explained by medical co-morbidities. Barriers to transplantation and to organ donation include: access to clear and understandable information, lack of cultural skills in relevant health providers, an over-emphasis on dialysis decision-making in the pre-dialysis period, and insufficient resources and operational support for patient workup. Year 1(2013) of the project predominantly involved the development of all the resources with years 2 to 4 of the project being the implementation period. Objectives The project aimed to increase live kidney donation offers and completed donation using a multi-faceted health service delivery improvement project to educate, encourage and facilitate live donation. We established a whole of renal service and community approach to improve live donation offers and transplantation. These included resource development, peer support, home based education, policy change, community engagement and primary health care education. •Live kidney donor offers increased from an estimated 10 per year prior to 2013 to over 80 per year from 2014. (Fig 1)•Live kidney donation has increased from an average of 2 per year prior to 2013 to over 10 a year from 2014. (Fig 2)•Number of living donor renal transplants increased in the Maaori and Pasifika groups from 5 (2 and 3 respectively) during 2006-2013 period to 17 transplants (6 and 11 respectively) during the period from 2013 to mid-2017.•Patient referral rates for renal transplant assessment and LKD increased from 39.4% to 59%.View Large Image Figure ViewerDownload Hi-res image Download (PPT) This project has demonstrated that a wide ranging approach to improving live kidney donation resulted in improved patient outcomes for priority populations.

Adherence to neonatal hypoglycaemia guidelines: A retrospective cohort study.

To determine if the routine use of automatically calculated birthweight centiles prior to discharge from the delivery unit is associated with improved adherence to the neonatal hypoglycaemia guideline. We conducted retrospective audits of adherence to the neonatal hypoglycaemia guideline in a tertiary maternity hospital in Auckland, New Zealand in a randomly selected cohort of newborn infants at risk of neonatal hypoglycaemia before (2011) and after (2015) the introduction of routine use of calculated birthweight centiles for all infants. The primary outcome was adherence to the guideline, defined as (i) blood glucose concentration screening in the first 48 h after birth; (ii) the initial measurement taken 1-2 h after birth; and (iii) at least three consecutive blood glucose concentrations ≥2.6 mmol/L, over 12 h, prior to cessation of screening. The audits examined the records of 400 infants (200 each in 2011 and 2015) to determine guideline adherence. Adherence improved from 2011 to 2015 (59/200 (30%) vs. 95/200 (48%), P < 0.001), with the largest improvement in large-for-gestational age infants (7/50 (14%) vs. 25/50 (50%), P = <0.001). Screened infants whose care was adherent to the guideline had a higher incidence of hypoglycaemia detection (adherent, 64/154 (42%) vs. non-adherent, 34/246 (14%), P < 0.001). The routine use of calculated birthweight centiles was associated with improved adherence to the neonatal hypoglycaemia guideline and increased detection of neonatal hypoglycaemia in at-risk infants. Thus, identifying practices that improve guideline adherence may improve the detection of hypoglycaemia in asymptomatic at-risk infants.

Impact of e‐scooter injuries on Emergency Department imaging

Since the introduction of a shared e-scooter service to Auckland in October there have been multiple media reports of associated injuries, but no quantitation of the number or severity of these injuries, or the impact on hospital emergency department services in Auckland. We performed a retrospective chart review on all patients referred to Auckland hospital ED radiology with the indication containing 'e-scooter' between 15 August 2018 and 15 December 2018. All requests were screened to ensure that the injury was caused by an e-scooter. Recorded data included patient demographics, type of imaging utilised, injury type, and whether admission or surgery was required. Sixty-four patients met the inclusion criteria, only one of these was prior to introduction of shared e-scooters on 15 October 2018. Of these, there were 27 limb fractures, 3 dislocations, a fractured spine, 12 patients with concussion, 1 extra-dural bleed, 9 facial/skull fractures and multiple soft tissue injuries. Almost 40% of the patients required admission to a specialty service following imaging, and 25.4% required surgery. A total of 221 plain films and 47 CT scans were performed for e-scooter injuries in the 2-month period after their introduction. Introduction of shared e-scooters has resulted in a large number of serious related injuries that have required urgent radiology imaging. Many of these patients required further specialist consultation or surgery, and place an increased burden on overstretched emergency department services.

Wound closure and follow-up after total knee arthroplasty — Do they affect the rate of antibiotic prescription?

BackgroundThis study aimed to evaluate risk factors for oral antibiotic prescription in the first six weeks after primary TKA, particularly whether the wound closure method (staples or sutures) and two-week follow-up clinician (surgeon or general practitioner (GP)) altered antibiotic use. MethodsFour thousand eight hundred forty-six TKAs from January 2013 to December 2016 at three tertiary hospitals in Auckland, New Zealand were analysed by manual review of patient electronic records and a national prescription database. Surgeon preference dictates the method of wound closure and whether wound review is followed up by the operating surgeon or by the patient's GP.Univariate and multivariate analysis was carried out to identify significant patient and surgical risk factors for oral antibiotic prescribing. ResultsOral antibiotics were prescribed in 24% of patients following primary TKA. Twenty-six percent of patients closed with staples were prescribed oral antibiotics versus 19% with sutures (adjusted OR = 1.4, p < 0.004). Excluding re-presentations and readmissions, GPs prescribed oral antibiotics in 22% of patients compared to seven percent of patients seen by surgeons (adjusted OR = 2.8, p < 0.001). Other risk factors for antibiotic prescription included increasing age, BMI and ASA score. ConclusionOral antibiotic prescribing rates are higher if the wound was closed with staples and if a GP performed the two-week follow-up. Improved communication between surgeons and GPs are required to ensure adequate follow-up following TKA and appropriate oral antibiotic use.

Computed tomography colonography: Retrospective comparison of laxative plus barium tagging versus iodinated contrast only for bowel preparation and faecal tagging.

There is no consensus as to the most optimal bowel preparation regime for Computed Tomographic Colonography (CTC). Concerns regarding laxative bowel preparation (LBP) include tolerability, adverse effects and complexity of regimen. Two hospitals in Auckland changed CTC bowel preparation from standard LBP to Gastrografin (GG) in 2015. The aim of this study is to retrospectively assess objective and subjective quality measures of these different bowel preparations. Two study groups were selected retrospectively from patients who underwent CTC at two hospitals in September-October 2013 (LBP) and September- October 2015 (GG). Each study group comprised 60 patients (30 consecutive patients from each hospital). Patients were randomized and anonymized to reduce bias in analysis. Study patients were assessed independently by three experienced Radiologists using a simple grading system derived from the literature. There was significantly less faecal residue (P=0.006) and better faecal tagging (P=0.001) in the right colon in the GG group. There was significantly higher fluid residue in the GG group than the LBP group (P=0.0001), particularly in the right colon, with better fluid tagging in the GG group(P≤0.0001). Higher Hounsfield Units of residual fluid were observed in the GG group (P≤0.0001). There was no statistically significant difference in the subjective quality scores (P=0.219), between the two preparations. This study has demonstrated better faecal cleansing and faecal tagging in the GG group. Higher fluid residue in the GG group was offset by better fluid tagging.

Reassessing the role of the human chorionic gonadotropin (hCG) discriminatory zone in the evaluation of patients with a pregnancy of unknown location

Objective This study aimed to determine the usefulness of the Human Chorionic Gonadotropin (hCG) Discriminatory Zone (DZ) (defined as the serum hCG value above which an intrauterine gestational sac is consistently seen on ultrasound in normal pregnancies1) as a diagnostic tool in the evaluation of patients presenting symptomatically (lower abdominal pain and/or vaginal bleeding) with a pregnancy of unknown location (PUL). A PUL is defined as a clinical scenario where a woman has a positive pregnancy test and no evidence of either intrauterine pregnancy or ectopic pregnancy on transvaginal ultrasound2. This objective will be achieved in this study by asking three questions: Firstly, how useful is the DZ in identifying an ectopic pregnancy? Secondly, how useful is the DZ in identifying a failed intrauterine pregnancy? And third, can a patient with an hCG level above DZ levels and a PUL following the initial scan can progress to a live, intra uterine pregnancy? Methods A retrospective chart review of ultrasound records at one, large urban hospital in Auckland, New Zealand identified 97 patients with lower abdominal pain and/or vaginal bleeding, who had serum hCG testing and transvaginal ultrasound performed within 12 hours of initial hospital presentation. The following information was recorded: hCG level, initial ultrasound findings, follow up outcomes, age, gravida and parity. Statistical analysis using Kruskal-Wallis and Wilcoxon were then performed. Results In total, 97 women met the criteria and were included in the study. Twenty-seven women had an hCG level of 1000 IU/L or higher, with true PUL on ultrasound. Of these, three went on to develop a normal intrauterine pregnancy, three were ectopic and 20 were miscarriages. The highest hCG value with a true PUL that progressed to a live intrauterine pregnancy was 2767 IU/L. The hCG discriminatory zone showed wide variability in hCG values and no statistically significant difference between median hCG values in the initial transvaginal ultrasound groups. The difference in median hCG values were found to be statistically significant between the final outcome groups of IUP, miscarriage and ectopic. However, the large overlap in values across the groups shows that hCG has low specificity. Conclusion This study recommends limited reliance on the hCG DZ and raises awareness of the high variability of hCG values. The hCG DZ was not shown to be useful in this study in the evaluation of women with a pregnancy of unknown location.