Infection is a potential complication inherent in surgery. When operation wounds are closed, 90% are contaminated by pathogenic bacteria. In order to prevent surgical wound infection, proper surgical techniques and the general state of the patient are the critical factors. The prophylactic administering of antibiotics has been found to help reduce the frequency and seriousness of infections when performed in the correct way .
The correct application of antibiotic prophylaxis should achieve adequate tissue levels of the antibiotic during what is considered the 'critical period', which lasts from 3 to 4 h from the moment of surgical incision . According to the Surgical Infection Society , this should comply with the following requisites:
(a) The first dose is to be administered prior to surgical incision, most typically during anaesthetic induction.
(b) Its duration should be of under 24 h.
(c) Use is to be made of the appropriate spectrum of antibiotics for the particular operation.
As elsewhere [2,3], Valladolid University Hospital has drawn up its own guidelines on surgical prophylaxis with antibiotics. However, it has been pointed out  that despite the periodical publication of surgical prophylaxis guides, the level of compliance is low.
The aim of this paper is to evaluate the level of compliance of the antibiotic prophylaxis according to the criteria of the Valladolid University Hospital guide.
A prospective study was designed and carried out during April 2003 in the surgery area of the Valladolid University Hospital, equipped with 800 beds. The study was approved by the Research Commission. Included in the study were all patients who, on Mondays and Wednesdays, underwent elective clean or clean-contaminated surgery and were admitted to hospital.
The surgeons responsible decided on the prophylactic antibiotics to be administered. The day before the operation, the surgeon wrote on the theatre sheet the prophylaxis protocol (antibiotic and dose), based on the hospital's guidelines on surgical prophylaxis. In the theatre, the nurse prepared the antibiotic, dissolving it in 50 mL of diluent and then asked the anaesthetist if it may be administered through the vein which has been canalized in preparation for the anaesthetic.
Data collection was undertaken by two anaesthesiologists on the day of the operation. Following surgery in each operating theatre, the clinical histories of the patients operated on were revised and the nurses and/or anaesthetists who had participated in the operation were asked to give further details.
Information concerning age, sex, type of surgery and antibiotic used was collected, and also whether the antibiotic had been prepared by the nurse before anaesthesia. Based on the hospital's guidelines on surgical prophylaxis, the appropriate prophylactic administration was considered to be:
(a) First dose prior to skin incision.
(b) Administration of a second dose in operations of over 2 h.
(c) Administration of antibiotic appropriate for the operative site.
(d) Administration of the proper dose.
A total of 323 patients, aged between 7 and 95, 59.1% male and 40.9% female, were analysed. The distribution of the patients according to surgical specialities was as follows: 10.8% cardiac, 16.7% vascular, 8.0% gynaecology, 6.5% neurosurgery, 15.2% orthopaedic, 14.5% general surgery, 7.7% ear, nose and throat, 8.7% urology, 7.1% thoracic and 4.6% plastic surgery.
A total of 14 antibiotics were used for prophylaxis: cefazolin, cefonicid, cefotaxime, ceftriaxone, vancomycin, amoxicilin-clavulanate, phosphomycin, netilmicin, gentamicin, mefoxitin, metronidazole, cefuroxime, tobramycin and clindamycin.
The surgical prophylaxis protocol was incorrectly observed (in accordance with the aforementioned criteria) in 190 patients (62.1%): the main reasons were the administering of the antibiotic following incision of the skin (70%) and non-administration of the antibiotic in interventions lasting over 120 min (30.5%, Table 1).
Of 306 patients analysed, in the case of 290 (91.9%), the theatre nurse had prepared the antibiotic prior to initiating anaesthetic induction.
The most relevant finding in our study is that in 190 (62.1%) of the patients antibiotic prophylaxis was incorrectly performed. This most frequently involved administration following the initiation of surgical intervention and the non-administering of a second dose of antibiotic in operations lasting over 120 min. Furthermore, in 91.9% of the cases, the antibiotic was prepared for administration before the initiation of anaesthesia and was, nevertheless, administered late.
Other authors [4,5] have likewise made reference to a high number of incorrect applications of antibiotic prophylaxis. Silver and colleagues , in a retrospective study carried out in 44 US hospitals, stated that 54% of the patients received no antibiotic within 2 h preoperatively. Gyssens and colleagues , in a university hospital in the Netherlands, found 68% non-compliance, the main reason being suboptimal timing in 53% of operations.
Inadequate surgical prophylaxis is possibly one of the factors that contribute to the high rate of surgical infection in Spanish hospitals and in the Valladolid University Hospital in particular which, for the year 2003, had values of 4.9% and 7.6%, respectively , noticeably higher than those documented in the literature for other similar centres [3,7]. Our study does not clarify whether this failure to meet the guidelines has an important impact on patient care.
Unlike other studies [4,5], in which compliance with antibiotic prophylaxis protocols was assessed retrospectively by means of data reviewing clinical histories, a fact which implies that a proper appraisal of the antibiotic used, the time of its administration and the dose is not undertaken, the prospective design of our study allows for a direct evaluation accurately reflecting the reality of the problem. However, we do not know if these data could be applied to other hospitals, either in our country or elsewhere.
In our opinion, the high rate of non-compliance with antibiotic prophylaxis protocols shows that the drawing-up of local guidelines (in this case without the participation of the anaesthetists) is insufficient to solve the problem. The lack of medical leadership, as regards decision-making on perioperative prophylaxis compliance, determines the need to assign responsibility in order to ensure proper application.
This explains why in our study, although in 91.9% of operations the antibiotic had already been prepared by the nurse, it was administered late. As the anaesthetists are responsible for administering drugs in the theatre, they only concentrate on the anaesthetics, forgetting about the antibiotic prophylaxis, possibly because they did not participate in the writing of the guidelines.
This highlights the suggestion that all involved departments (including both surgery and anaesthesia) should actively audit their own antibiotic prophylaxis guidelines and practices, and instigate changes, where appropriate, to improve surgical prophylaxis protocols.
We propose as a viable solution that the anaesthesiologist, as the doctor in charge of administering drugs in the operating theatre, also be given responsibility for surgical prophylaxis and be allowed to play an active part in establishing the guidelines. In addition, these activities should be complemented by a periodic revision of these local guidelines together with annual prospective monitoring of the rate of compliance.
On the other hand, we are aware of the negative attitudes generated by this idea in the fields of both anaesthesia and surgery in Spanish hospitals. Gyssens  has already stated that to achieve a proper administration of antibiotics, the anaesthesiologist must be involved and that the times of administration must be noted down on the anaesthesia record.
Department of Anaesthesiology and Reanimation; Valladolid University Hospital; Valladolid, Spain
Department of Medicine and Public Health; Faculty of Medicine, University of Valladolid; Valladolid, Spain
F. J. Álvarez
Department of Pharmacology and Therapeutics; Faculty of Medicine, University of Valladolid; Valladolid, Spain