Reduction of the Single Breath CO Diffusing Capacity in Cystic Fibrosis

Reduction of the Single Breath CO Diffusing Capacity in Cystic Fibrosis· D.]. Cotton, M.D., F.C.C.P.; B. L. Graham, Ph.D.;]. T. Mink, B.Sc.; and B. F. Habbick, M.D.

The single breath diffusing capacity of the lung for carbon monoxide (Dsb) was measured using three equations to describe CO uptake separately during inhalation, breath holding, and exhalation in 24 patients with cystic fibrosis and 30 control subjects with similar age and height distributions. Using the control group, we developed two prediction equations for Dsb: one based on height, age, and sex; and another based on alveolar lung volume (VA) to the 2/3 power. based on We also developed a prediction for DsblVA (~:()) height. The Dsb as percent predicted (% pred) using either

I n cystic fibrosis,

pathologic studies indicate the presence of panacinar emphysema, 1 bronchiolar dilatation," and a relative reduction in capillary number and increase in pulmonary arterial wall thickness with advancing right ventricular hypertrophy," Despite evidence of pulmonary vascular hypoplasia':' and loss of elastic recoil' in this disease, the single breath diffusing capacity of the lung for carbon monoxide (Dsb) has been reported to be either increased," decreased," or unchanged."?" Since recent pathologic studies suggest involvement of the pulmonary microcirculation':" in this disease, and since it has been recognized that conventional methods":" of measuring Dsb may be affected by alterations in exhaled flow rates,":" as well as the size and timing of the alveolar gas sample,":" we have re-examined the changes in Dsb in cystic fibrosis using a new method":" that *From the Section of Respiratory Medicine, Department of Medicine and Department of Pediatrics, College of Medicine, University of Saskatchewan, Saskatoon, Canada. This work was supported by the Canadian Cystic Fibrosis Foundation, the John Moorhead Foundation, the MRC of Canada (MA-7053), and the Saskatchewan Lung Association. Manuscript received April 10; revision accepted July 10. Reprint requests: Dr. Cotton, ,544 Ellis Hall, University II osp ita I, Saskatoon, Saskatchewan. Canada S7N OXO

prediction equation decreased with increasing age and height as well as with decreasing % pred maximal midexpiratory Row rate (FEF25-75) in cystic fibrosis patients but not in controls. The Keo (% pred) also decreased in cystic fibrosis with increasing age and decreasing percent pred FEF25-75. We conclude that in patients with cystic fibrosis, Dsb decreases with variables that relate to increasing disease severity (age, height, and increasing airflow obstruction).

minimizes these effects. We have found that Dsb decreases in cystic fibrosis with increasing age, height, and airflow obstruction. METHODS

We studied 24 patients with cystic fibrosis and 30 control subjects. After obtaining informed consent, all filled out a standard technician-administered questionnaire." The diagnosis of cystic fibrosis was made by standard clinical criteria" and by the presence of an elevated sweat chloride (Table 1). The patients had not experienced any acute changes in symptoms in the previous six weeks. The control subjects were recruited through the cooperation of hospital employees. All control subjects were free of acute respiratory symptoms in the previous six weeks, denied any previous history of respiratory symptoms by questionnaire, were lifetime nonsmokers, and denied a history of any other chronic medical condition. Subjects and patients performed maximal expiratory flow-volume maneuvers, and patients also performed single breath nitrogen tests using equipment previously described." To measure the Dsb, we recorded flow and volume with a pneumotach mounted on the wall of a bag-in-box system and measured instantaneous changes in carbon monoxide (CO) and helium (He) concentration during single breath maneuvers using rapid responding analyzers for CO and He. 16•17 Both patients and control subjects were studied seated at rest with a nose clip in place. Each subject and patient performed two acceptable trials of a single breath maneuver which consisted of slow inhalation and exhalation (10 percent) of the vital capacity/second) and either 5 or 10 seconds (s)of breath holding at TLC. We calculated

Table I-Anthropomorphic and Pulmonary Function Data (Mean±l SD)

Sex

n

M/F

Control

30

17/13

Cystic Fibrosis

24

14/10

Age (yr)

Height

13.3 ±4.0 14.2 ±3.6

156 ± 16 154 ± 15

(ern)

Weight (kg) 46.7

± 14.3 41.5 ± 12.2

TLC % of pred

FVC % of pred

FEV 1 % of pred

FEF25-75 % of pred

I)sh ml/rnin/ mm Hg 10 sec breath hold

103.0 ± 12.2 94.9* ±11.9

95.7 ± 12.7 84.5* ± 18.4

97.1 ±11.1 66.2* ±21.8

95.0 ±20.8 38.3t ± 2.5.9

20.3 ±6.2 18.3 ±4.9

*p<0.05. tp
217

Dsb using three separate equations to describe CO uptake during the single breath maneuver":" and obtained alveolar volume from analysis of the mean He concentration in the exhaled vital capacity. 17 We have previously found" that this method provides a better estimate of lung volume in patients with airflow obstruction than previously employed single breath dilution methods. Analyses of variance and co-variance were used to test for equality of slopes. stepwise regression analysis to determine prediction equations and linear regression analysis to assess correlations between Dsb and anthropomorphic variables as well as other tests of lung function. Unpaired two-tailed Students t-test was used for comparison of control subjects vs cystic fibrosis patients. Values (Table 1)obtained from control subjects and cystic fibrosis patients were compared using the following previously published normal regressions: for the forced expired volume in one second (FEV,). the forced vital capacity (FVC). and the maximum rnidexpiratory flow rate (FEF25-75) from Knudson et al;20 for total lung capacity (TLC) from Weng and Levison." and for the phase 3 slope of the single breath nitrogen test (4N/L) from Cooper et aI. 22 This study had been approved by the University of Saskatchewan Presidents Advisory Committee on Ethics in Human Experimentation.

, -.70 lIe.G01

1 o

40 eo eo FEF 25-75 (l\: 01 'UD I

...-......... 100

I

120

..•...

·

Sl,00

RESULTS

218

20

_ 120

f

The mean age, height, weight, and male to female ratio were similar between control subjects and cystic fibrosis patients (Table 1) and the distributions of age and height were also similar. The FVC, FEV!> MMFR, and FEV/FVC x 100 (p<0. (01) and TLC (p<0.05) were all lower in patients than in control subjects (Table1). For the patient group, the hemoglobin value was 14.0±1.5 g/dl (mean z I SD) and the sweat chloride level was 112 ± 25 mEq/L (n = 23). In normal subjects, Dsb for the 10 s breath holding maneuver correlated with height (r = 0.88; p
....-...--.

0 ....

. r'.M

.'.001

o

20

40

fEF 25-75

eo

f% 01

eo

100

120

'UD'

...... •

~12O

SlIOO

:

.. eo

, '.12 • '.001

8

o

20

.'0

80

80

fEf 25-75 (% 01 'RED)

I

100

FIGURE 1. Correlation between % pred Dsb from equation 1 (upper graph). equation 2 (middle graph). and % pred Kco (lower graph) vs % pred FEF25-75 (regression obtained from control subjects). Solid circles are individual data and the solid line is the linear regression for cystic fibrosis patients. For comparison, the linear regressions for the data from the control subjects (dashed line) (p =0.5 or greater) is shown in each graph.

Dsb (% pred) and FEF25-75 (% pred) when Dsb was predicted from age, height, and sex (Equation 1, Fig 1). A similar correlation using equation 1 was found for Dsb (% pred) vs % pred" FEV/FVC x 100 (r= .56; p<0.(05), in cystic fibrosis patients but not in control subjects. The Dsb (% pred, equation 1) also declined with increasing % pred ~N/L (r= - .67, p
140

•

= cr

•

• • - .61 P

120

·.001

•

I

S f

100

"

~80

• ••

•

"0

i

60

•

,

10

,

12

,

(y.)

AGE 140

•

cr

• I

S

...'"

60

i

60

'0

•

•

••

'0

~

• • -.57 p·.004

• ---------.-------

100

18

16

•

-;:; 120

,

;

14

•

• • •

,

10

•

12

AGE

i

14

20

• : -.56 P ".005

•

f

,

(y.)

• t'l

• • •

18

16

140

__ 120

•

•

• •

• 80

o t,.. .~I

o

,

' . . . . . . . . . . . . . . . . . .- - - . . - - - . . - - - . - - - - . - - - - . -

8

10

,

18

---.

I

20

In patients with cystic fibrosis (but not in control subjects), the Kco (% pred) declined with decreasing % pred FEF25-75 (Fig 1). We found that while there was no decline in Dsb (% pred, equation 1) with age or height in control subjects, Dsb declined 2.8 percent of expected per year of age (Fig 2) and 0.6 percent of expected per centimeter increase in height (r= - .54; p<0.OO5) in cystic fibrosis patients. The Dsb expressed as percent 2 predicted using VA / 3 (Equation 2) declined 2.3 percent of expected per year of age (Fig 2) and declined. 47 percent of expected per centimeter increase in height (r= - .50; p=O.Ol) in cystic fibrosis patients but not control subjects. The Keo declined 2.2 percent of expected per year of age (Fig 2) and declined. 46 percent of expected per centimeter increase in height (r= - .49; p<0.05). In 11 patients with a low FEV, (48.4 ± 10.0 % pred) compared to the 13cystic fibrosis patients with a higher FEV 1 (81.2± 17.1 % pred), we found that Dsb (% pred) using either equation 1 or 2 and Kc:o (% pred) were all significantly lower (p<0.05) in the most vs the least obstructed group (Table 2). Although we have compared Dsb in control subjects vs cystic fibrosis patients using data from a 10 s breath holding maneuver, we have also measured Dsb following 5 s of breath holding. For 5 s of breath holding, Dsb was significantly lower (p<0.05) in cystic fibrosis patients (17.3±4.8 ml-min t'smm Hg- 1) compared to control subjects (20.9±6.6 ml-min t'smm Hg- 1) . The Dsb following 5 vs 10 s of breath holding was similar in control subjects but was decreased in cystic fibrosis patients (p<0.05). Furthermore, Dsb was lower in cystic fibrosis patients vs control subjects following 5 s of breath holding using analysis ofco-variance with either age alone (p<0.05) or height alone (p<0.05) as the independent variable. Although TLC was lower in cystic fibrosis patients than in control subjects (Table 1), we found no difference in the measured TLC for the 5 s breath hold (3.96 ± 1.36 L) vs the 10 s breath hold maneuvers (4.03 ± 1.37). When we examined the changes in % pred TLC in cystic fibrosis patients, we also found no significant correlations with age, height, FEF25-75 (% pred), Keo (% pred), or Dsb (% pred, equation 2), but Dsb (% pred, equation 1) decreased significantly with decreasing TLC (p<0.05). DISCUSSION

control subjects. When Dsb was expressed as a ratio of the simultaneously measured volume, this ratio (Keo ) declined with height (r= -.4; p<0.05) in control subjects: Equation 3: Ke o=6.8- .014· Ht (ern)

In this study, we have found that Dsb decreases significantly with increasing age, height, and airways obstruction in patients with cystic fibrosis compared to a group of normal children of similar age, height, and sex distribution. This finding supports previous observations on changes in the pulmonary vasculature that occur in this disease.'? At autopsy,':" cystic fibrosis CHEST I 87 121 FEBRUARY. 1985

219

Table 2-Dsb in Cystic Fibrosis Patients with Lower vs Higher FEV, (% PredJ (Mean±l SD)

FEV.

s60% pred

Dsb

% Pred

Keo % Pred

TLC (% Pred, 22)

(Equation 1)

(Equation 2)

(Equation 3)

13

81.2 ± 17.1

99.0 ±7.4

99.2 ± 14.7

106.7 ± 12.8

107.3 ± 13.0

11

48.4+ ± 10.00

90.1 ± 14.6

81.5t ± 14.1

92.6* ±12.7

94.7* ±12.9

n

FEV. >60% pred

Dsb

% Pred

FEV. (% Pred, 21)

·p
patients with right ventricular hypertrophy (RVH) had reduced background haze by arteriography, fewer arteries per unit area of lung, and increased intra-acinar arterial wall thickness proportional to the degree of RVH. Based on the response of the pulmonary vasculature to generalized hypoxia in rats," Reid' postulated that hypoxemia in cystic fibrosis might cause premature muscularization of intra-acinar pulmonary arteries and lead to a reduction in intra-acinar artery number. Since small airways obstruction" and pulmonary hypertension" may develop early in cystic fibrosis, the normal growth of the pulmonary vasculature may be uniquely affected because normal alveolar multiplication continues up to age eight and alveoli continue to increase in diameter and develop new intra-acinar arteries" until skeletal growth ceases. The progressive reduction in Dsb with advancing disease could be caused not only by failure of alveolar multiplication and/or enlargement, but also by failure of normal development of the pulmonary capillary bed. The observed reduction in Dsb in cystic fibrosis might have been due to reductions in hemoglobin (Hb) with advancing disease, but no correlation was found whatsoever between Hb and FEF25-75 (% pred) in our patients (p<0.5). While previous studies using body plethysmographic methods suggest that TLC in cystic fibrosis was normal," it was found that TLC is slightly lower in cystic fibrosis patients compared to our normal subjects. It is possible that, with increasing nonuniformity of ventilation, the single breath helium dilution method of measuring alveolar volume could have spuriously underestimated alveolar volume, and hence, could have lowered the estimation of Dsb in cystic fibrosis patients with the greatest airflow obstruction. 12 When we predicted Dsb on the basis of VA 2/3 or examined Keo (predicted on the basis of height), however, we found that both Keo and Dsb still decreased with increasing airflow obstruction indicating that the reduction in Dsb and K e o with advancing disease is not due solely to the influence of disease on the size of the lungs or on underestimation ofTLC but is due, at least in part, to changes in the pulmonary 220

microcirculation. In the present study, six cystic fibrosis patients (25 percent) had a Dsb of80 percent or less when predicted on the basis of age, height and sex. When we compared control subjects with cystic fibrosis patients using standard reference physical growth statistics," we found that the cystic fibrosis patients were significantly shorter (p<0.05) for their age (40th ± 30th percentile) than control subjects (60th ± 30th percentile). This indicates that predicted Dsb values based on height in cystic fibrosis would also be lowered by the influence of the disease on skeletal growth which would mask reductions in the % pred Dsb. Finally, there was no difference in TLC from the 5 vs 10 s breath holding maneuver in cystic fibrosis patients, but Dsb was significantly lower for the shorter breath hold time. During lung growth in control subjects, Dsb may be influenced by several factors. Since our control subjects were seven to 21 years old, lung growth may be due primarily to an increase in the size of existing alveoli in this group. Under these conditions, lung surface area, and hence, Dsb might increase with VA 21J • 6.27.28 In this study, we have confirmed previous observations" that Dsb correlates strongly with VA213. When growth is complete in adults, the alveolar number is quite variable but appears to correlate best with body length." Based on this observation, Dsb should also correlate strongly with height, a relationship found by us in the present study and previously by others." Finally, O'Brodovich et al" have pointed out that in the sitting position, Dsb may be influenced by skeletal growth in normal children because, with increasing height, perfusion becomes reduced in a progressively increasing amount of lung in the upper lung zones. This phenomenon causes a progressive derecruitment of apical pulmonary capillaries in the sitting position with increasing height so that Dsb becomes progressively lower in the apex vs the base. In the normal adult seated subject, Dsb appears considerably lower in the apex vs the base. 31 The difference in Dsb between the supine and sitting position, which increases with body height in normal children, reflects the increasing nonuniformity of difSingleBreathCO DiffusingCapacityin CF (Cotton st 8/)

fusion that exists in the sitting but not the supine position. While we did not study the effect of body po-

sition in this study, others have observed?" that Dsb

fails to increase in the supine vs sitting position in patients with cystic fibrosis. These observations suggest that prediction values in control subjects for Dsb would be progressively greater in the supine vs sitting position with increasing height. The differences that we found in the sitting position between control subjects and cystic fibrosis patients might be even more marked in the supine position. In a previous study, Dsb was increased in a group of cystic fibrosis patients and was also higher in those patients most severely obstructed." The discrepancy between the results in this study and the previous study of Keens et al," may in part be due to errors in Dsb measurements that occur using conventional methods to calculate Dsb. Conventional methods of measuring Dsb overestimate Dsb in normal subjects when the exhaled flow rate is reduced":" and in patients with reduced exhaled flow rates caused by airflow obstruction." For example, in normal adult subjects, we found that Dsb calculated by the method of Ogilvie et al'' increased 17.5 percent when the exhaled How rate was reduced from 2 Us to 0.5 Us. In contrast to conventional methods, we have used three separate equations," one each for inhalation, breath holding, and exhalation to calculate Dsb in this study. U sing this method, we have previously reported that Dsb is not affected by alterations in exhaled flow rates both in a lung model" and in normal adult subjects. 16 Finally, in the same group of cystic fibrosis patients taking part in this study, we have reported that conventional methods overestimated Dsb compared to the three-equation method. 17 Because the three-equation method does not introduce errors in Dsb calculations with reduced inhaled and exhaled flow rates either in normal subjects or patients with airflow obstruction, we were not restricted to the "standardized'?' forced inhalation, 10 s breath hold, forced exhalation single breath maneuver. We chose single breath maneuvers with slow inhaled and exhaled flow rates because cystic fibrosis patients could perform the maneuver in the same manner as the control subjects and because this maneuver would reduce the possible influence of widely fluctuating pleural pressures" and improve the distribution of inhaled gas in patients with airways obstruction. 33 We intentionally chose to analyze the 10 s breath hold maneuver in greater detail because previous investigations in COPD patients" indicate that Dsb using the three-equation method decreases with decreasing breath hold time. We also found that Dsb was lower in cystic fibrosis patients for the 5 s vs the 10 s breath holding maneuver in the present study. It has been suggested that the decrease in Dsb with decreas-

ing breath holding time in patients with ventilation maldistribution may reflect diffusion limitation within the airways due to impaired mixing of inhaled gas with resident gas." While the cause of the increase in Dsb with increasing breath hold time in the presence of airflow obstruction remains speculative, we have examined the differences between normal individuals and cystic fibrosis patients with long breath hold times where potential differences would be minimized. Although Dsb was similar in control subjects for 5 s vs 10 s of breath holding (r = 95; p<0.00(1), the differences between control subjects vs cystic fibrosis patients are actually greater using 5 s vs 10 s breath holding maneuvers in this study. In summary, we conclude that cystic fibrosis causes a reduction in Dsb, and that this reduction correlates with the progression of the disease as measured either by increasing age or height, or by worsening forced exhaled flow rates. ACKNOWLEDGMENTS: We thank Mrs. R. Day for preparing the manuscript and Mrs. S. Patola for her administrative assistance. REFERENCES

1 Reid L. Cardiopulmonary pathology. In: Perspectives in cystic fibrosis-Proceedings of the 8th International Congress on Cystic Fibrosis. Mississauga: Imperial Press, 1980: 198-214 2 Esterly JR, Oppenheimer EH. Cystic fibrosis of the pancreas: structural changes in peripheral airways. Thorax 1968; 23:670-75 3 Reid L. The pulmonary circulation: remodelling in growth and disease-The 1978 J Burns Ambertson Lecture. Am Rev Respir Dis 1979; 119:531-45 4 Mansell A, Dubrowsky C, Levison H, Bryan AC, Crozier ON. Lung elastic recoil in cystic fibrosis. Am Rev Respir Dis 1974; 109:190-97 5 Keens TG, Mansell A, Kratins RB, Levison H, Bryan AC, Hyland RH, et al, Evaluation of the single-breath diffusing capacity in asthma and cystic fibrosis. Chest 1979; 76:41-44 6 O'Brodovieh HM, Mellins RB, Mansell AL. Effects of growth on diffusion constant for carbon monoxide. Am Rev Respir Dis 1981; 125:670-73 7 Russel NJ, Bagg LR, Hughes DTD, Neville E. Lung function in young adults with cystic fibrosis. Br J Dis Chest 1982; 76:35-43 8 Beier FR, Renzetti AD, Mitchell M, Watanabe S. Pulmonary pathophysiology in cystic fibrosis. Am Rev Respir Dis 1966; 3: 430-40 9 Baran D, Englert M. Pulmonary mechanics and diffusing capacity in cystic fibrosis. Helv Paediat Acta 1973; 28:175-82 10 Zelkowitz PS, Giammona S1: Effects of gravity and exercise on the pulmonary diffusing capacity in children with cystic fibrosis. J Pediatr 1969; 74:393-98 11 Ogilvie eM, Forster RE, Blakemore WS, Morton JW. A standardized breath holding technique for the clinical measurement of the diffusing capacity of the lung for carbon monoxide. J Clin Invest 1957; 36:1-17 12 Ferris BG, ed. Epidemiology standardization project: recommended standardized procedures for pulmonary function testing. Am Rev Respir Dis 1978; 118:62-72 13 Jones RS, Meade F: A theoretical and experimental analysis of anomalies in the estimation of pulmonary diffusing capacity by the single breath method. Q J Exp Physioll961; 46:131-43 14 Cotton OJ, Newth CJL, Portner PM, Nadel]A. Measurement of the single breath CO diffusing capacity by continuous rapid CO analysis in man. J Appl Physiol: Respirat Environ Exer Physiol CHEST I 87 I 2 I FEBRUARY, 1985

221

1979; 46:1149-56 15 Graham BL, Dosman]A, Cotton DJ. A theoretical analysis of the single breath diffusing capacity for carbon monoxide. IEEE Trans Biomed Eng 1980; 27:221-26 16 Graham BL, Mink J1: Cotton DJ. Improved accuracy and precision of single breath CO diffusing capacity measurements. J Appl Physiol: Respirat Environ Exer Physioll98L 51:1305-13 17 Graham BL, Mink J1: Cotton DJ. Overestimation of the single breath diffusing capacity in patients with airflow obstruction. Am Rev Respir Dis 1984; 129:403-08 18 Wood RE, Boat TF, Doershuk CE State of the art-cystic fibrosis. Am Rev Respir Dis 1976; 113:833-78 19 Cotton DJ, Graham BL, Li K-YR, Froh F, Barnett GD, Dosman ]A. Effects of smoking and occupational exposure on peripheral airway function in young cereal grain workers. Am Rev Respir Dis 1982; 126:660-65 20 Knudson RJ, Slatin RC, Lebowitz MD, Burrows B. The maximal expiratory Row-volume curves. Am Rev Respir Dis 1976; 113: 587-600 21 Weng T-R, Levison H. Standards of pulmonary function in children. Am Rev Respir Dis 1969; 99:879-94 22 Cooper DM, Mellins RB, Mansell AL. Changes in distribution of ventilation with lung growth. J Appl Physiol:Respirat Environ Exer Physioll98l; 51:699-705 23 Rabinovitch M, Gamble W, Nadas AS, Miettinen OS. Reid L. Rat pulmonary circulation after chronic hypoxia: hemodynamic and structural features. Am J Physiol 1979; 236:818-27 24 Lamarre A, Reilly BJ, Bryan AC, Levison H. Early detection of pulmonary function abnormalities in cystic fibrosis. Pediatrics 1971; 50:291-98

222

25 Goldring RM, Fishman A~ Turino GM, Cohen HI, Denning CR, Andersen D H. Pulmonary hypertension and cor pulmonale in cystic fibrosis of the pancreas. J Pediatr 1964; 65:501-24 26 Hamill P~ Drizd TA, Johnson CL, Reed RB, Roch AF, Moore WM. Physical growth: National Center for Health Statistics percentiles. Am J Clin Nutr 1979; 32:607-29 27 Ogliati R, Levine D, Smith J~ Briscoe W. King TKC. Diffusing capacity in idiopathic scoliosis and its interpretation regarding alveolar development. Am Rev Respir Dis 1982; 126:229-34 28 Weibel ER. Morphological basis of alveolar-capillary gas exchange. Physiol Rev 1973; 53:419-95 29 Angus GE, Thurlbeck WM. Number of alveoli in the human lung. J Appl Physiol 1972; 32:483-85 30 Bucci G, Cook CD, Barrie H. Studies of respiratory physiology on children: V. Total lung diffusion, diffusing capacity of pulmonary membrane and pulmonary capillary blood volume in normal subjects from 7 to 40 years of age. J Pediatr 1961; 58: 820-28 31 MacIntyre NR, Nadel JA. Regional diffusing capacity in normal lungs during slow exhalation. J Appl Physiol: Respirat Environ Exer Physiol 1982; 52:1487-92 32 Cotton DJ, Mink J1: Graham BL. Effect of high negative inspiratory pressure on single breath CO diffusing capacity. Resp Physiol 1983; 54:19-29 33 Engel LA, Macklem IT Gas mixing and distribution in the lung. In: International review of physiology respiratory physiology II. Baltimore: University Park Press, 1977:37-82 34 Cotton DJ, Mink J1: Graham BL. Effect of breath holding time on DLcoSB in patients with airflow obstruction. Physiologist 1981; 24:91

Single Breath CO Diffusing Capacity in CF (Cotton et 8/)