1.gif (1892 bytes)

Brief Reports

Indian Pediatrics 2002; 39:1039-1043

An Indigenously Developed Method for Sweat Collection and Estimation of Chloride for Diagnosis of Cystic Fibrosis


S.K. Kabra, M. Kabra, S. Gera, R. Lodha, Sreedevi KN, S. Chacko,
J. Mathew, S. Shastri and M. Ghosh

From the Department of Pediatrics, All India Institute of Medical Sciences, New Delhi 110 029, India.

Correspondence to: Dr. S.K. Kabra, Additional Professor, Department of Pediatrics, AIIMS, New Delhi 110 029, India.

E-mail: [email protected]

Manuscript received: March 7, 2002; Initial review completed: April 5, 2002;

Revision accepted: May 31, 2002.

 

An indigenously developed method for sweat collection and titration method for estimation of chloride was validated. The mean difference in estimated chloride value from the known strength of saline in 50 samples was –1.04 ± 4.13 mEq/L (95% CI: –0.07 to 2.28). The mean difference in the estimated chloride values between two observers when the test was performed on known strengths of saline solution was –2.5 ± 4.24 mEq/L (95% CI: –3.67 to 1.33). The inter observer variability between two observers when the test was performed on sweat samples obtained from 50 individuals was –1.12 ± 4.34 mEq/L (95% CI: –2.23 to 0.8 ). Sweat weight of more than 100 mg could be collected in first attempt in 602 of 757 (80%) patient with an average sweat weight of 230 mg. This inexpensive method of sweat collection and chloride estimation has acceptable accuracy and repeatability and can be used in resource poor setting for making a diagnosis of cystic fibrosis.

Key words: Cystic fibrosis, Low cost technology, Sweat test.


Demonstration of elevated levels of chloride in sweat is the most commonly used test for diagnosis of cystic fibrosis(1). Sophisticated equipments are available for estimation of sweat chloride. However, these are not readily available in resource poor countries due to high initial and on-going costs. Due to non-availability of the sweat chloride estimation, diagnosis of cystic fibrosis is often missed(2). To overcome this problem, we developed a simple method for sweat chloride estimation. In this method, sweat was collected by pilocarpine iontophoresis and chloride was estimated by performing titration. In this communication we present the details of method along with its validation.

Method

The method was developed in Pediatric Pulmonology division of All India Institute of Medical Sciences, New Delhi.

Sweat collection equipment: It was based on the Gibson and Cooks method(3). Six batteries of 1.5 volts each were connected in series. The electric wires were connected to positive and negative terminals respectively. ECG electrodes were connected to distal ends of each wire for pilocarpine iontophoresis.

Procedure of sweat collection: The flexor aspect of forearm was cleaned with distilled water and dried with clean swabs. Two pieces of filter (Whatman filter paper no 42 size 2.5 Χ 2.5 cm) soaked in 1% pilocarpine solution were placed under positive electrode and the electrode was secured over the flexor aspect of the forearm with the help of velcro. Similarly flexor aspect of arm in the same limb was cleaned and 5 filter papers soaked in plain water were placed under negative electrode and it was secured on the arm with velcro. The completion of circuit was checked from reading of multimeter. The iontophoresis procedure was continued for 5-7 minutes. The value of maximum current was recorded from multimeter. If the observed current value was less than 1.5 mA, iontophoresis was repeated. After 5-7 minutes of iontophoresis, electrodes were removed and skin under the positive electrode was cleaned with distilled water and dried with swabs. Filter paper from the pre-weighed bottle (bottle weighed with filter paper) was taken out with the help of forceps and placed over the cleaned skin. The filter paper was secured firmly to skin with the help of a cling film. After 30 minutes filter paper was removed and put back into the bottle. The bottle along with filter paper containing sweat was weighed on an electronic weighing scale with an accuracy of 0.1 mg. The weight of the collected sweat was calculated by the difference in weight (before and after sweat collection). The minimum sweat weight that could be used for titration was 100 mg. If the sweat weight was less than 100 mg, iontophoresis was repeated. The sweat in filter paper was eluted in 8 mL distilled water over– night(12-16 hours).

For assessment of tolerance and adverse reaction, sweat was collected on 10 volunteers (doctors).

Sweat chloride estimation: Sweat chloride was estimated by Schales and Schales method(4). It included titration on 2 milliliters of eluted solution with 0.002N mercuric nitrate using diphenyl carbazone as indicator. The amount of mercuric nitrate required for titration was recorded in milliliters. Titration was repeated with 0.01N saline. The chloride was calculated by following equation N1 V1 = N2 V2. The simplified formula from this equation was as follows:

8+sweat weight (g) Test volume (mL)

Chloride (mEq/L) = ——————Χ5Χ ——————

sweat weight (g)Χ2 Volume with

saline (mL)

Validity of the method was tested by performing chloride estimation on different strengths (10/samples each of chloride concentration 154, 77, 51, 30 and 15 mEq/L) of saline with different weights. The mean and standard deviation of the difference from the standard was calculated and 95% confidence interval was estimated.

For assessment of reliability the test was assessed by two observers (SKK, SC) on 50 samples containing known strengths of saline in different weight and 50 sweat samples collected from patients sent for sweat test. The mean and standard deviation of the difference in chloride values between two observers and 95% confidence interval was calculated.

Results

Using the equipment we collected sweat in 10 volunteers (doctors) without any adverse reaction.

Validation of the method: The mean difference between estimated chloride value and the known strength of saline in 50 samples was –1.04 ± 4.13 (95% CI : –0.07 to 2.28).

Reliability of the method: The mean difference between estimated chloride value between two observers when the test was performed on known strength of saline solution ( n = 50) was –2.5 ± 4.24 mEq/L (95% CI: –3.67 to 1.33). The inter-observer variability between two observers when the test was performed on sweat samples obtained from 50 patients sent for sweat chloride estimation was –1.12 ± 4.34 mEq/L (95% CI: –2.23 to 0.8). The sweat chloride value of > 60 mEq/L was observed in 12 (24%) and 14 (28%) by the two observers respectively. If we take sweat chloride value of more than 60 mEq/L as abnormal then two samples (4%) were misclassified. The sweat chloride values of less than 40 mEq/L was observed in 32 (64%) and in 34 (68%) samples by two observers respectively. If we consider a sweat chloride value of less than 40 mEq/L as normal then 2 samples (4%) were misclassified. The four misclassified values were in the borderline range.

Cost of the equipment was 7 US$ (INR 350) and cost per test was 0.5 US$ (INR25).

Experience with the above mentioned method: Over a period of 5 years sweat was collected on 757 patients. 531 were males. Mean age of the patients was 6.16 years (1 month to 45 years). Sweat of more than 100 mg could be collected in 602 (80%) patients in the first attempt. In 155 patients we failed to collect sweat weighing more than 100 mg in first attempt. Out of these 155 patients sweat collection was done during months of October to March in 95 (61%). 113 out of 155 patients came for sweat testing second time and we could collect sweat (>100 mg) in 84 (74%). 10 of the 27 patients (in whom adequate sweat could not be collected after two attempts) came for sweat testing third time and in 4 adequate sweat could not be collected. Average sweat weight was 230 ± 87 mg (95% CI: 222 to 240). During October to February average sweat weight was 211 ± 82 mg (95% CI: 192 to 228) while the same during March to October was 244 ± 88 mg (95% CI: 233-256). There was no significant difference in the failure rate in sweat collection in relation to sex or age of the patients. The maximum current value recorded was 3.0 mA. The side effects of the method included minor skin burns at the site of negative electode in 10 patients in the first 6 months.

Discussion

In the indigenously developed method of sweat collection we could collect sweat weighing more than 100 mg in 80% of the patients in the first attempt. The success rate of 75-95% with various methods has been described in literature(5-16). The common side effect reported in methods using quantitative pilocarpine iontophoresis are skin burns in 1-10% of the patients at the site of electrodes(17,18). We observed minor skin burns in 10 (1.3%) of the patients. They were because of damaged negative electrode. This side effect did not occur after the electrodes were replaced and we used 5 layers of filter paper. The failure rate of sweat collection was more during winter months and the average sweat volume was also less during these months. This was because of the low ambient temperature of 10-20 degree C. We believe that due to cold weather there was significant vasoconstriction and the effective sweat collection was less due to decreased blood supply to the sweat glands. If a patient can be provided a controlled environment the failure rate during these months may be reduced.

We used titration method for estimation of chlorides. The accuracy and inter personal variation were within acceptable range. Non-agreement between two observers for sweat chloride values according to the accepted cut off values (normal < 40 mEq/L and abnormal >60 mEq/L) was observed in 8% of the samples. However, none of the normal values reported by one observer was classified as abnormal by the other observer. The same was true for abnormal values. The misclassification can be reduced if patients with abnormal or borderline values are subjected to another estimation of sweat chloride, which is the practice at our center.

The chances of under/over estimation of chloride in this method may be due to improper cleaning and drying of skin, evaporative loss of water from sweat collected on filter paper during transfer from skin surface to bottle, improper cleaning of test tubes or pipettes leaving some chloride ions in them and subjective error in recognizing change in color at the time of titration. With proper precautions and experience these errors can be minimized. It is recommended that only an exprienced person should perform the sweat test even when automatic equipments are being used(1).

We conclude that this simple method of sweat collection and chloride estimation may be an alternative to other expensive methods where cost is a limiting factor for making this investigation available.

Contributors: SKK developed the equipment, planned the study, standardized the method, collected sweat, and prepared manuscript and will act as guarantor for the paper, MK helped in making the equipment, standardizing the method and writing the manuscript. SG helped in standardization of the equipment. SC, SKN and JM helped in sweat collection and standardization of the method. RL helped in writing the manuscript. SS and MG helped in preparing reagents for sweat test and in preparing manuscript.

Funding: The method was developed as a part of project "To establish cystic fibrosis services in Delhi" funded by International Cystic Fibrosis (Mucoviscidosis) Association.

Competing interests: None stated.

Key Messages

• A simple and inexpensive method of sweat collection based on quantitative pilocarpine iontophoresis gives reasonable rate of sweat collection and chloride values can be estimated by titration. This method can be an alternative to expensive equipment in resource poor setting.

 

 References


REFERENCES

1. LeGrys VA. Sweat testing for the diagnosis of cystic fibrosis: practical considerations. J Pediatr 1996; 129: 892-897.

2. Kabra SK, Madhulika, Connett GJ, Rolles CJ. Diagnosis of cystic fibrosis - Indian perspective. Indian J Pediatr 1999; 66: 923-928.

3. Gibson LE, Cooke RE. Test for concentration of electrolyte in cystic fibrosis of pancreas utilizing pilocarpine by iontophoresis. Pediatrics 1959; 23: 545-549.

4. Schales O, Schales SS. A simple and accurate method for determination of chloride in biological fluids. J Biol Chem 1941; 140: 879-848.

5. Warwick WJ, Hansen LG, Brown IV, Laine WC, Hansen KL. Sweat chloride: quantitative patch for collection and measurement. Clin Lab Sci 2001; 14: 155-159.

6. Webster HL, Quirante CG. Micro-flow cell conductometric sweat analysis for cystic fibrosis diagnosis. Ann Clin Biochem 2000; 37: 399-407.

7. Pogorzelski A, Zebrak J, Tomalak W. The usefulness of the conduction sweat test in diagnosis of cystic fibrosis. Pediatr Pol 1995; 70: 639-543.

8. Warwick WJ, Hansen LG, Brown I. Improved correlation of sweat chloride quantification by the CF Indicator System and the Gibson-Cooke Sweat Test. Clin Chem 1993; 39: 1748-1754.

9. Riedler J, Arrer E. Comparison of the classical Gibson-Cooke methods and the chloride-sensitive electrode in sweat testing for diagnosis of cystic fibrosis. Pediatr Pathol 1991; 26: 173-175.

10. Simmonds E, Alfaham M, Prosser R, Penney MD. Fractional measurement of sweat osmolality in patients with cystic fibrosis. Arch Dis Child 1989; 64: 1717-1720.

11. Sprossig C, Wunderlich P, Bergmann S, Hoffmann C. Diagnosis of mucoviscidosis by examination of sweat: value of conductometry. Kinderarztl Prax 1989; 57: 497-502.

12. Barnes GL, Vaelioja L, McShane S. Sweat testing by capillary collection and osmometry: suitability of the Wescor Macroduct System for screening suspected cystic fibrosis patients. Aust Pediatr J 1988; 24: 191-193.

13. Northali H, York GA. Sweat Sodium and chloride analysis using BM/Hitachi 911 ion selective electrodes. Brit J Biol Sci 1995; 52: 68-70.

14. Denning CR, Huang NN, Cuassay LR, Shwachaman H, Tocci P, Warwick WJ, et al. Cooperative study comparing three methods of performing sweat test to diagnose cystic fibrosis. Pediatrics 1980; 66: 752-757.

15. Hammond KB, Turcios NL, Gibson E. Clinical evaluation of the macroduct sweat collection system and conductivity analyzer in the diagnosis of cystic fibrosis. J Pediatr 1994; 124: 255-260.

16. Varlotta L. Management and care of the newly diagnosed patient with cystic fibrosis. Curr Opin Pulm Med 1998; 4: 311-318.

17. LeGrys VA, Retsch-Bogart GZ. Urticaria associated with the pilocarpine iontophoresis sweat test. Pediatr Pulmonol 1997; 24: 296-297.

18. Rattenburry JM, Worthy E. Is the sweat test safe? Some instances of burns. Ann Clin Biochem 1996; 33: 456-458.

Home

Past Issue

About IP

About IAP

Feedback

Links

 Author Info.

  Subscription