This retrospective study evaluated variables associated with length of stay (LOS) in hospital for 406 admissions of primary cleft lip and palate and alveolus surgery between January 2007 and April 2009. Three patients were treated as day cases, 343 (84%) stayed one night, 48 (12%) stayed 2 nights and 12 (3%) stayed >2 nights. Poisson regression analysis showed that there was no association between postoperative LOS and age, distance travelled, diagnosis and type of operation, with a p value >0.2 for all variables. 60/406 patients stayed 2 nights or more postoperatively mostly due to poor pain control and inadequate oral intake. Patients with palate repair were more likely to have postoperative LOS >1 night, compared to patients with lip repair, p value = 0.011. Four patients (1%), all of whom had undergone cleft palate surgery, were readmitted within 4 weeks of the operation due to respiratory obstruction or haemorrhage. Using logistic regression, evidence showed that these readmissions were related to a longer original postoperative LOS. This study shows that length of stay for primary cleft lip, palate and alveolus surgery can in most cases be limited to one night postoperatively, provided that adequate support can be provided at home.
Length of stay (LOS) in hospital for primary cleft lip and palate (CLP) surgery and alveolar bone grafting (ABG) varies considerably worldwide, depending on local conditions and traditions, health care systems, geographic areas, transportation systems and the opinion of the surgeon in charge. The decision to limit LOS should primarily be governed by patient factors such as safety and comfort, while ensuring that treatment standards are not compromised. International studies have shown that it is the surgeon’s personal approach as well as local factors influencing their decision that may lead to variations in LOS. Westert et al. reported that LOS variations exist between different hospitals where multihospital physicians performed comparable surgical procedures on comparable groups of patients. De Jong et al. carried out a similar study in New York and concluded that physicians adapt to their colleagues’ or to the management protocols of the hospital in which they work.
In the UK, in a recent report, the National Health Service claimed that length of hospital stay is one of the greatest variables between NHS trusts, but with simple measures, hospitals can improve the patient experience by reducing the number of days spent in hospital, increase capacity and save money. It is predicted that lengthy hospital stay will become exceptional and by shortening the LOS, hospitals will benefit from a reduction in costs as well as greater patient satisfaction, without seeing an increase in the rate of readmissions. By introducing a more personalized care plan through trained specialist nurses, relatives and friends, it is thought that patients may have a faster recovery time in the comfort of their own homes. Contrary to this, staying in hospital for longer may provide a more rapid response to a postoperative emergency and access to repeat surgery. The latter is seldom likely to be required for most CLP surgery, including bone grafting, except in patients with Pierre Robin sequence who spend the first night in a high dependency unit for observation of their breathing patterns and oxygen saturation.
There are a number of examples of reductions in LOS for various cleft surgical procedures. Lees and Pigott advocated a 3 day postoperative stay after cleft lip repair as they found that respiratory and haemorrhagic complications arose within 48 h. They suggested a 5 day stay for cleft palate repairs as there was significant morbidity in this group until the fifth postoperative day, although 4 in 7 of these patients had a known history of pulmonary problems. Anthony and Sloan addressed these concerns and found that airway obstruction often occurs within 2 h postoperatively; the risk in the first 24 h is 4.4% and thereafter it is minimal. Cronin et al. conducted a retrospective study of cleft palate operations carried out over a 10 year period and found that a 1 night stay is safe for healthy patients undergoing routine cleft palate surgery. Of the 79 cases they reviewed, 94% stayed for 1 postoperative day and 97% stayed less than 1.5 days. They concluded that each case should be considered individually, and ultimately, the decision for discharge lies with the surgeon.
The South Thames Cleft Service is based at the Evelina Children’s Hospital (ECH) as part of Guy’s and St Thomas’ NHS Foundation Trust. It is one of the 11 regional cleft services in the UK. It covers the rural and metropolitan regions of Kent, Surrey, Sussex and the southern half of greater London ( Fig. 1 ), a population of nearly 7 million, with an incidence of 130 newborn babies with CLP defects per year. The authors hypothesize that a one night stay postoperatively should be safe and adequate for primary CLP repair as well as ABG. A prolonged LOS for patients should be the result of an associated complex medical history or social factors. The aims of this study are to evaluate retrospectively the preoperative and postoperative LOS of patients with primary CLP repairs and ABG operations at the ECH within a set time frame, and to investigate the following variables in relation to postoperative LOS: age, diagnosis, operation type, surgeon and distance to the hospital.
Materials and methods
The South Thames Cleft service covers a large geographic region so patients and their carers are offered a stay in hospital accommodation the night before surgery. As part of the management of patients undergoing cleft surgery, early postoperative home care is encouraged. Ward rounds are performed routinely before and after surgery and the following morning by the surgical team and a cleft nurse specialist (CNS). The CNS authorizes discharge on condition that there are no problems with the surgical site, the patient has an adequate oral intake, pain is controlled, patient observations are stable, there are no respiratory problems and that there is a stable home environment to which the patient can return. The parents are provided with good postoperative home care instructions and a review appointment 6 weeks postoperatively, but a CNS also visits the patient at home during the first postoperative week and liaises with the surgeon if there are any concerns. The parents can phone the CNS at any time day or night if they are worried or if there is an emergency.
A retrospective study was carried out on all patients who had undergone primary CLP repairs and ABG surgery from January 2007 to April 2009 at the ECH. Data were collected from operation diaries and the hospital’s electronic patient records (EPR) data bank, together with diagnosis and surgical details extracted from the hospital’s cleft service spreadsheet. Any outstanding information including reasons why patients stayed >1 night or why they were readmitted was obtained from the clinical records.
406 surgical episodes were identified. The patient’s age, diagnosis, procedure, surgeons, preoperative and postoperative LOS, distance travelled and time taken by car to the hospital, readmissions and reason for >1 night postoperative stay were recorded.
Four surgeons operated on this patient group and they were labelled anonymously as A, B, C and D. The diagnoses included unilateral cleft lip (UCL), bilateral cleft lip (BCL), unilateral cleft lip and alveolus (UCLA), unilateral cleft lip and palate (UCLP), bilateral cleft lip and palate (BCLP) and cleft palate (CP). The operations carried out were primary lip and palate repairs and ABG with cancellous or cortico-cancellous iliac crest bone. The exclusion criteria for this study included a diagnosis of submucous cleft repair, secondary cleft procedures as well as more than one procedure simultaneously. One patient was excluded because of incomplete clinical records.
The patients’ home postcodes were found in the EPR, then entered into the UK website ‘AA route finders’ to determine the estimated distance from the patient’s home to St Thomas’ Hospital. Lees and Piggot have suggested that two factors that may influence a patient’s LOS, are the distance from the patient’s home to the hospital and access to transport. Therefore, the relationship between postoperative LOS and estimated distance travelled was analysed.
The data were collected, transferred into Microsoft Access and subsequently Microsoft Excel and SAS v9.2 were used to calculate summary statistics and formulate tables, graphs and draw results from the data set. Summary histograms were produced for independent variables of interest; patient’s age (years), distance travelled to the hospital (km), diagnoses and type of operation. These were also produced for postoperative LOS (number of nights).
Poisson regression was used to examine the association between postoperative length of stay and age, estimated travel time, diagnoses and type of operation. Initially univariable models were fitted to examine each variable separately. Variables having a p value <0.2 were included in a forward step-wise multivariable model in order to adjust for confounding effects. Logistic regression was used to examine the association between readmission and postoperative LOS. Fisher’s exact test was used to find out if patients with palate repairs had significantly higher prolonged LOS compared to patients with lip repairs.
38% of the operations were lip repairs, 46% were palate repairs and 16% were ABG. The mean age for patients undergoing lip repair, palate repair and ABG surgery at ECH was 3.6 months, 1.0 year and 10.9 years respectively ( Table 1 ). Figure 2 shows a distribution of how far patients travelled from their home to the hospital.
|Age for lip repair (years)||153||0.3||0.3||0.25||3.50|
|Age for palate repair (years)||188||1.0||0.7||0.25||1.08|
|Age for alveolar bone graft (years)||65||10.9||10.5||1.00||16.83|
Table 2 shows that the mean total LOS was 1.2 nights, ranging from a minimum of 0 nights (day case) to a maximum of 7 nights. This ranges from 0 (day case) to 7 nights for lip repairs, 1–6 nights for palate repairs, and 1–2 nights for ABGs. 15 (4%) patients were admitted 1 day preoperatively. Postoperatively, 3 (1%) patients were day cases, 343 (84%) stayed one night only, 48 (12%) stayed 2 nights and 12 (3%) stayed more than 2 nights ( Fig. 3 ).
|Total stay (nights)||406||1.2||1.0||0||7|
|LOS for lip repair (nights)||153||1.14||1.0||0||7|
|LOS for palate repair (nights)||188||1.26||1.0||1||6|
|LOS for ABG (nights)||65||1.12||1.0||1||2|
Results from Poisson regression analyses of postoperative LOS showed that the p values for all variables (estimated distance, age, diagnosis and type of operation) were >0.2 and therefore not statistically significant, meaning no multivariable model was constructed ( Table 3 ).
|Variable||Level||Rate ratio||95% CI||Overall p value|
|Estimated distance (km) *||1.00||(1.00, 1.00)||0.983|
|Age **||0.99||(0.97, 1.02)||0.617|
|Lip repair||0.89||(0.73, 1.07)|
LOS was 0–2 nights for 100% of patients who had ABGs, 95% of patients who had lip repairs and 96% of patients who had palate repairs. There is evidence that the proportion of patients whose LOS is >1 night is higher in patients with palate repair compared to lip repair ( p = 0.01). A higher proportion of patients having palate repair had >1 night postoperative LOS, compared with lip repair and ABG operations ( Table 4 ).