Address for Correspondence:

Dr P.V Ramachandran, 
Professor of Radiodiagnosis, 
Kerala Health Research Welfare Society
Imageology Centre, Medical College, Calicut- 673008
Calicut, Kerala
E-Mail: pvramachandran@yahoo.com  

ABSTRACT

Introduction: Few studies have been conducted on Computerised Tomography(CT) scans in a large number of cases , drawn from a variety of departments. Here an attempt is made to correlate clinical diagnoses with the CT findings.
Objective: To study the CT positivity in 5020 consecutive cases and determine its characteristics
Methodology: The study was conducted at the Kerala Health Research Welfare Society Imageology Centre attached to the Medical College, Calicut. A retrospective analysis of 5020 consecutive cases was carried out and variables related to mode of referral and the results of CT study were studied to determine the usefulness of CT.
Results: The CT diagnosis was the same as clinically suspected in 60%. Cases from surgical specialities, emergency cases and contrast studies were found to be less contributory to clinical management than medical specialities, routine cases and non-contrast cases respectively. Active clinical - radiological discussions and communication of clinical scenario and previous investigation results contributed significantly to CT positivity.
Conclusion: CT is an expensive mode of investigation with attendant risks and judicious use should be made of it. More stringent criteria need to be set for CT referral. 
Keywords: Computerised Tomography,Retrospective Analysis, Usefulness, Diagnosis

INTRODUCTION

The patients referred for a CT scan can be divided into three groups:
1) Where CT is mandatory- acute trauma, bowel pathology, parenchymal lung disease, staging solid tumors and lymphoma, treatment planning, etc. 
2) Where CT is indicated but equal diagnostic accuracy is possible by alternative modalities- as in imaging of solid abdominal organs- where USG and MRI can be used and vascular disorders where DSA, Doppler and MRI are alternatives. 
3) Where CT can be performed but other modalities are definitely superior and not done due to non availability or other reasons e.g. imaging of spinal cord, IV disc and white matter disorders where MRI has greater diagnostic accuracy and in gall bladder calculi where USG is sufficient.
In groups 1 & 3 there is little doubt as to the usefulness of CT as a diagnostic modality, but in group 2, careful evaluation of the positivity of CT studies is essential. We must give more attention to group 2 patients and weigh the risks associated with CT and its financial considerations against the benefits of doing the CT study. Often management decisions and diagnostic protocols may need to be altered. The cases where other modalities may be superior to CT should be avoided with the following beneficial results

 
· To reduce time spent on such (group 2) patients and save scanner time for group 1 patients 
· To reduce workload on scanner staff and thus improve diagnostic accuracy.
· To reduce the financial burden on patients and hospitals in general as well as society as a whole.

An average number of 15 CT cases are done per day in our department. The average cost of one CT study to the patient is Rs.1000/- in our unit which is the minimum charge compared to other privately owned CT Scanners. When we consider that an average middle class family in Kerala earns about Rs.3000/- to Rs.4000/- per month we can imagine the financial impact that a sophisticated diagnostic tool like CT has on the patients and therefore, on society. The cost of Radiological procedures is a major contributor to the cost of health care. The economic impact of this is therefore a matter of great concern in a developing country like India. All over the world, Radiological procedures are often overused with unproductive outcomes. It is not possible to make perfect patient selection for diagnostic radiology. Therefore any attempt to predict the usefulness of radiological procedures in different clinical settings is a complex task.

Auditing is an essential component of intelligent resource management. Clinical audit should be viewed as " a scrutiny of clinical performance, policies and practice with a view to improving patient care" 

This retrospective study of 5000 computed tomographies undertaken , is part of auditing in our center. By this study we hope to prove the usefulness of CT as a diagnostic modality. The basic components of audit include structure , process and outcome. Structure is what we need - resources - space, staff, equipment and technology. Process is what we do - efficiency of staff - by assessing diagnosis and treatment of patients. Outcome is what we achieve. Auditing has a positive impact by 

1. Improving patient care 
2. As a tool of learning
3. For optimizing resources
4. Improving efficiency and 
5. Reducing risks to patients
Doctors should realize that they should have internal controls within their own profession for self-appraisal and to minimize scrutiny and criticism by non-medical persons. 


AIM OF THE STUDY

To study the CT positivity in a series of 5000 consecutive cases and determine its characteristics.


MATERIALS AND METHODS


CT is an important modality for diagnosing diseases. It gives greater diagnostic accuracy than conventional X-rays. It is estimated that there are about 1900 X-ray installations all over Kerala and 5% of X-ray installations in the State are CT units which while offering highly beneficial clinical information also cause considerably large radiation hazard to the patients. There are about 96 CT installations in Kerala and the number is fast increasing with the accelerated sale of units by all leading manufacturers. According to the status report of Directorate of Radiation Safety, Government of Kerala, it is found that only 43 institutions out of these 96 CT installations satisfy radiation safety regulations.(1) All CT examination requests should be screened by the Radiologist to ensure that CT is the appropriate modality for detection of this specific clinical problem.

The Kerala Health Research Welfare Society (KHRWS) Imageology Centre is a facility for Radiodiagnostic imaging attached to the Medical College - Calicut. The hospital is a tertiary level referral center for five districts in North Kerala. The total bed strength exceeds 3000 with all major specialties and superspecialities. An average of 4500 CT studies are conducted at our center annually.

1. Selection of subjects
Inclusion criteria - consecutive 5000 patients who underwent CT examination in
the KHRWS Imageology Centre. Only technically successful 
cases were included
Exclusion criteria - Those cases where adequate clinical details were not available.
.
2. Data Collection
Demographic, clinical and laboratory data as detailed in the proforma were collected. A full time research assistant abstracted requisition forms and information from the department record library. Data accumulated in the study were entered in the computer and results tabulated.

A retrospective study was done with 3168 male patients and 1852 female patients. The number of paediatric patients (below the age of 16) was 643. Two CT scanners were used:

CT unit No.1 Shimadzu SCT 3000 Tx - conventional CT 3rd generation 
(250 MA 300MAS). Here tube and detectors are positioned on opposite sides of a ring that rotates around the patient. The physical linkages between the power cables mean that the tube is unable to rotate continuously. After each rotation the scanner stops and rotates in the opposite direction. Nowadays conventional scanners are retained alongside new scanners. They have a role - especially in non contrast examinations that do not require fast scanning for optimal vascular enhancement. A large bulk of CT workload - routine examinations of head for cerebral infarcts or hemorrhage- gets done on the conventional scanner; thus freeing time on the spiral scanner.

CT Unit No.2 Shimadzu SCT 7000 Ts Spiral CT Imager (160 MA 300MAS) Spiral scanners (developed in the late 1980s ) use slip ring technology, eliminating the need for a rigid mechanical linkage between power cables and X ray tube. While the tube is rotating, the table supporting the patient also moves continuously so that a volume of tissue rather than individual slices is scanned. The data are then reformatted automatically to display the images as axial slices. High quality reconstructed images in coronal, sagittal and oblique planes can be obtained.
Advantages
· Short scan time
· Quality reconstructions from closely spaced scans
· Different phases of contrast enhancement can be studied
· Likelihood that a small lesion may be overlooked is less
The variables included in the study were referring department, type of requests, clinical diagnoses, previous investigations, part examined, CT diagnoses, any suggestions to the clinician given in the CT report.

The subjects had no pre-procedural preparation except fasting for at least 4 hours in contrast examinations and they had normal hydration status. Patient position was optimized - tailored to each patient part scanned. 
In most patients the CT examination time lasted for less than thirty minutes. 
Methods of analysis
All data were entered in prestructured and pretested proforma. Appropriate descriptive statistics were used to summarize clinical data, that is, means and standard deviations for continuous variables and proportions for nominal variables. A level of less than 0.05 was fixed for the 'P' value for inferring statistical significance. Analysis was done using SPSS 10. Chi-square was used to determine associations between the variables.



RESULTS 

The sample comprised of 5020 individuals with majority of them male (n=3168. 63%). Sixty two percent of the subjects (n=3101) were referred from medical specialties and 38% (n= 1906) were from surgical specialties. The age distribution of the subjects were as follows: 12.8% below 16 years, 19.6% were aged 16 to 30 years, 18.7% were in the 31 to 45 years age group, 23% from 46 to 60 years, and 25.9% were aged more than 60 years. Majority of the requests were for routine scans (n= 3506, 70%), and 21% were requests for emergency scans (n=1061). Follow up studies comprised of 9% (n=453) of the total scans. Based on the clinical diagnoses the distribution was such that intracranial pathologies predominated. Head injuries - 1141 (22.7%), CVA - 1251 (24.9%) and ICSOL - 612 (12.2%) formed the largest groups. Intracranial bleed, CNS infections, hydrocephalous, disorders of the Para nasal sinuses, musculoskeletal, chest and abdominal pathologies and others formed small groups (<10% each). The results of previous investigations were studied and the properly investigated cases, effectively communicated, were 1500 (29.9%). The presenting conditions were themselves revealing or were actively discussed in 3511 (69.9%) cases. The remaining 9 (0.2%) cases had not been investigated. The parts examined were head and neck in the majority - 4161 (83.1%). Thorax was examined in 474 (9.4%) instances and abdomen in 190 (3.8%). PNS, spine limbs and others constituted the remaining small minority. Contrast was not administered in 3207 (63.9%) and was used in 1813 (36.1%) of patients. 

The CT diagnosis was the same as that clinically suspected in 2981 (59.4%). It was not the same as clinically suspected but explained the clinical scenario in 244 (4.8%). No significant pathology was detected in 1795 cases (35.8%). Repeat CT was advised in 112 (2.2%) patients only. Other imaging modalities were suggested in 183 (3.6%) of cases. Usefulness of CT was compared across the medical and surgical specialties and it was found that in 67.5% of the medical and 58.9% of the surgical specialty cases, CT was useful in management.( Chi square=38.235 , p= .000).Even when follow up cases were excluded the differences remained significant - 67.2% vs 55.8 %.(Chi square= 57.525, p= .000). In the follow up cases there was no significant difference in the usefulness of CT - 74.8% vs 72.7% , medical vs surgical. We compared the usefulness of contrast and non-contrast CT and found a significant difference between the two. 58.9% of the contrast CT scans were useful while 67.2 % of the non contrast CT scans were useful.(Chi square = 33.017, p= .000). 3.6% (n=183) of the patients were advised other investigations. Interestingly only 15% of these cases were those in which CT did not contribute to clinical management. When routine and emergency scans were compared only 50% of emergency scans were useful in clinical management compared to 67.4% of routine scans (Chi square = 106.768, p= .000). Cases which were properly investigated and the results were properly communicated to the radiologist showed a high CT positivity of 80.7%. Those conditions where the presenting conditions themselves were revealing however were positive in only 57.2% of cases.

DISCUSSION

The current study is important in that this is the first large-scale study that looks at the profile of CT scans carried out at a multi-specialty hospital. The advent of this useful radiological tool has resulted in enhancement of diagnostic process and also management aspects in almost all fields of medicine.(2) Studies conducted in UK have shown that General Practitioners were also found to be making judicious use of CT.(3). Conjunct to this has been a necessary evil of over investigation especially when cases are subject to CT scans without an adequate indication or when alternative methods are available. 

It is interesting to note that CT diagnosis was the same as that suspected clinically in majority of the scans (59.4%). Therefore the slightly increased initial expense (to do a CT) is made up by getting greater information to clinch the diagnosis. This would translate into the case being subject to a good clinical work up before investigation. Also, the probable lack of conclusiveness of clinical work up is indicated in the proportion of cases (4.9%) in which the CT diagnosis was not the same as that clinically suspected, but did help to explain the clinical scenario. These findings are comparable to other studies of estimating diagnostic accuracy.(4) But what is more important from the radiological perspective is that in 35.8% (n=1795) cases, there was no significant contribution from the CT study towards clarifying the clinical situation. This is a disappointing finding in a resource poor setup like India, if this proportion of cases implies unwanted investigations. Studies have been conducted over the past few years and they have found a significant increase in the use of CT and MRI in most studies with a different set of indications for CT as compared to MRI.(5) 
The relation between usefulness of the CT study and other variables indicate that there was a significant difference in the CT study with respect to the specialty from which requested. 

The requests were from various departments. The medical specialities were grouped together and included, general medicine, radiotherapy, pediatrics, physical medicine and rehabilitation, psychiatry, chest diseases/respiratory medicine and neuromedicine .The indications for CT & its usefulness were found to be similar in children when compared to adults.(6) The surgical specialties included general surgery, pediatric surgery, urology, ophthalmology, orthopedics, gynaecology, thoracic surgery, dental surgery, plastic surgery, ENT and neurosurgery. Those patients referred by general practitioners were grouped as a 3rd category.

The comparison of the usefulness of CT across the specialties, even though crude, offers some insight into beneficially modifiable practices.(7) The finding that surgical CT studies were less useful compared to medical may be because of a multitude of reasons. One possible explanation is that many cases in surgical specialities are referred for imaging on the benefit of doubt, e.g in cases of head injury. Strict adherence to the guidelines (deteriorating neurological status) for carrying out investigations strictly will help in reducing the unwanted burden of ineligible cases. A set of objective criteriae- Constitutional Signs and Symptoms (CSS) for head injury to screen patients receiving head trauma need to be used before using costly work up.(8) The integration of Radiologists in diagnostic and therapeutic algorithms significantly improves the outcome of patients with head injury. The time to establish a diagnosis has a crucial impact on good outcome of patients.(9) 

Patients were categorized based on their clinical diagnoses. The CT scans of the head and neck region constituted the majority. These included intracranial hemorrhage, head injuries, CVA, ICSOL, CNS infections, hydrocephalous, cortical vein thromboses, AVM, encephalocele, meningocele, orbital, tooth & socket, neck and base of skull pathologies.(10) The other clinical diagnoses included - pathology of the musculoskeletal system, chest, PNS, abdomen and spine. The studies of the pelvis where a preliminary ultrasonogram is indicated needs special attention. Here radiation risk could be reduced if the radiologist has an idea from the ultrasonogram as to the nature of the pelvic pathology.(11) Also those CT scans of the thorax exposing the breasts in women need definite indications after taking into account the radiation risk.(12)

The use of contrast in imaging studies is highly useful , at the same time fraught with risks. Hence it would be expected that contrast study subjects would be well chosen. It is in this context , that the finding of reduced positivity among contrast studies compared to routine plain studies, becomes significant. The fact that nearly 40% of the contrast studies turned out to be normal with little contribution to clinical management needs to be addressed. It would merit further studies to look into choosing patients for contrast studies with more stringent criteria. It is significant as the contrast studies cost approximately twice that of plain studies, and with double the radiation exposure. A prospective study(13) showed that a reduced radiation dose could be used especially in non contrast scans without compromising image quality.

Thus each poorly chosen contrast study would mean loss of opportunity to do two 'possibly' useful plain studies. 
The types of requests were grouped as emergency, follow up and routine scans.
We found that the usefulness of emergency scans were significantly lower than the routine scans.(14) This would seem to imply that there needs to be a better scrutiny of emergency cases and avoid over investigating just because of emergency. The most significant aspect in the current study is the finding that there is a significant correlation between thorough assessment and discussion with the radiologist prior to scan and the CT positivity. This is understandable, as imaging studies need to be in tandem with the clinical team to make use of its full potential. The radiologist is at a handicap when there is ineffective or no communication from the clinical team. Hence we suggest that in all cases, even when emergency, a thorough evaluation of the clinical situation and communicating the clinical scenario to the radiologist be encouraged to increase the CT positivity.(15) The advantages are two fold: this would allow the clinician to choose cases for imaging more judiciously and also offers the radiologist a better vantage point to interpret the radiological findings.
The major strength of the study is the large number of cases and the variety of departments from which patients were referred. Since the analysis was carried out by an independent person the possibility of bias was eliminated.

The limitations of the study include difficulty in follow up. Follow up was possible only if the patient was re-attending the center for a repeat CT or if a guided biopsy was carried out - in which case tissue diagnosis could be obtained from the pathology department. Only a brief clinical history was available. Because of this the clinical perspective may not have been evident and the study was predominantly from a radiological viewpoint.


CONCLUSIONS

CT is an expensive mode of investigation with the risks of radiation exposure (and of contrast injection in indicated cases). Hence judicious use should be made of CT. In the light of the findings of the current study we conclude that more stringent criteriae need to be set for CT referral. We also stress the need for replicating the study in other settings. Active clinical - radiological correlation is essential to increase the CT positivity.



References:

1. Report Submitted by Directorate of Radiation Safety, Kerala to AERB, 2003
2. Jane Hawanaur. Clinical Review on Recent Advances in Diagnostic Radiology, BMJ 1999; 319: 168-171.
3. Patick M Strong, Robert J Walker. General Practitioners also use open access computed tomography Wisely. BMJ 1995; 311: 325-326.
4. Robert Harper, Barnaby Reeves. Reporting of precision of estimates for diagnostic accuracy : a review. BMJ 1999; 318: 1322-1323.
5. Bunn HJ, Pugh RE, Thomson A. How has imaging of the head, neck and spine changed over 5 years in a district general hospital ? Pediatr Radiol 2002; Feb; 32(2): 110-3, Epub 2001.
6. Haydel MJ, Shembekar AD. Prediction of intracranial injury in children aged five years and older with loss of consciousness after minor head injury due to nontrivial mechanisms. Ann Emerg Med. 2003; Oct.42 (4): 507-14.
7. Conall J Garvey, Rebecca Hanlon. Computed Tomography in Clinical Practice. BMJ 2002; 324: 1077-1080.
8. Falimirski ME, Gonzalez R, Rodriguez A, Wilberger J. The need for head computed tomography in patients sustaining loss of consciousness after mild head injury. J Trauma 2003 Jul;55(1): 1-6.
9. Rieger J, Linsenmaier U, Pfeifer KJ, Reiser M. Radiological diagnosis in acute craniocerebral trauma. Radiologe 2002 Jul.42(7): 547-55.
10. RA Schoenenberger, SM Heim. Indication for computed tomography of the brain in patients with first uncomplicated generalized seizure. BMJ 1994; 309:986-989.
11. Mindelzun, Jeffrey. Unenhanced helical CT for evaluating acute abdominal pain. Radiology 1997; 205: 43-47.
12. Madan M Rehani, Manorma Berry .Radiation doses in computed tomography.. Radiation doses in computed tomography. BMJ 2000; 320: 593-594.
13. David Spurgeon. Radiation dose in scans could be halved - News Roundup. BMJ 2001; 323:185
14. Miranda Voss, John D Knottenbelt, Margaret M.Peden. Patients who reattend after head injury : a high risk group. BMJ 1995; 311: 1395-1398.
15. Paul Dorman, Peter Sandercock. Access to Computed Tomography in British Accident and Emergency Departments. BMJ 1997; 314: 400.





Source of Funding
The research project was undertaken with a funding provided by Atomic Energy Regulatory Board (AERB) , Govt. Of India through a project namely "AERB SR/18/01" sanctioned by the AERB Committee for safety research programmes ( letter no.AERB/SRP/18N/01/99/2351 dated 21May 1999) 

© Copyright of articles belongs to the respective authors unless otherwise specified.Verbatim copying, redistribution and storage of this article permitted provided no restrictions are imposed on the access and a hyperlink to the original article in Calicut Medical Journal maintained. All opinion stated are exclusively that of the author(s).
Calicut Medical Journal upholds the policy of Open Access to Scientific literature.