The quantitative lung index (QLI): a gestational age–independent sonographic predictor of fetal lung growth




Objective


We sought to develop a gestational age–independent sonographic parameter to characterize lung growth.


Study Design


Reported descriptors of lung growth, including lung-to-head circumference (HC) ratio (LHR) and observed/expected LHR, were examined. A new index, the quantitative lung index (QLI) was derived using published data on HC and the area of the base of the right lung.


Results


Neither the LHR nor the observed/expected LHR proved to be gestational age independent. Right lung growth can be expressed using the following formula: QLI = lung area/(HC/10)^2. The 50th percentile of the QLI remained constant at approximately 1.0 for the gestational age between 16-32 weeks. A small lung (<1st percentile) was defined as a QLI <0.6.


Conclusion


Fetal right lung growth can be adequately described using the QLI, independent of gestational age. Further studies are needed to assess the clinical accuracy of the QLI in characterizing fetal right lung growth.


Congenital diaphragmatic hernia (CDH) remains one of the most challenging conditions in fetal therapy. In utero approaches through open fetal surgery were initially aimed at repairing the defect but gave way to palliation of the condition in utero via tracheal occlusion (TO). Soon, open fetal surgery TO was itself replaced by percutaneous TO via direct laryngoscopic methods. In 2003, a randomized clinical trial conducted in the United States on TO via open fetal surgery vs expectant management for CDH did not show a benefit of therapy vs expectant management, with therapy-associated premature rupture of membranes in 100% of patients. Despite these perceived negative results, which suggested that expectant management was no different than in utero surgery, subsequent studies have shown a high incidence of neurological morbidity in both groups, suggesting that expectant management is not as advantageous as assumed. Thus, efforts to improve the outcome of CDH fetuses have continued. Attention has focused on identifying those fetuses for which in utero therapy would be of potential benefit.




See Journal Club, page 577



In 1996, Metkus et al proposed the use of the lung-to-head circumference (HC) ratio (LHR) as a sonographic measure to identify fetuses with CDH that had a poor prognosis. The LHR uses an assessment of right lung area (LA), measured at the level of the 4-chamber view of the heart. The longest diameter and the longest perpendicular diameter are multiplied and the product is divided by the HC. The LHR was assumed to be gestational age independent. The LHR was used as the basis for the 2003 US randomized clinical trial on TO via open fetal surgery vs expectant management for CDH. The criterion for entry in that trial was an LHR <1.4. In retrospect, the data suggested that a more restrictive LHR of <1.0 might have been more appropriate, since survival in this subgroup of CDH fetuses was only 33% compared to virtually 100% of both surgical and expectantly managed groups with an LHR >1. In a recent study, 210 CDH fetuses with an LHR <1.0 that underwent TO with a detachable balloon had a 49% survival, compared to an expected survival rate of 25% estimated from a regression model.


Notwithstanding the link between the LHR and neonatal survival, the use of the LHR as a sonographic criterion for surgical intervention criteria for CDH has come into question. Peralta et al, in a study of 650 normal fetuses from 12-32 weeks, demonstrated that the LHR increases exponentially with gestational age. As such, they suggested instead using percentiles of LA to discriminate between normal and small lungs. Jani et al proposed the concept of observed/expected (o/e) LHR to predict mortality in CDH, believing this ratio to correct for gestational age. However, the o/e LHR had only a 46% sensitivity in predicting perinatal mortality from CDH, with an attendant 10% false-positive rate. Other approaches, including 3-dimensional ultrasound and fetal magnetic resonance imaging have also failed to discriminate the at-risk populations.


Our group has worked on alternative sonographic methods to assess fetal lung growth.


The purpose of this paper is to describe the derivation of a sonographic mathematical index to quantify fetal lung growth independent of gestational age.


Materials and Methods


First, we examined the relationship of the LHR and the o/e LHR with gestational age. Percentiles of right LA ( Table 1 ) and for the LHR ( Table 2 ) were derived from the article of Peralta et al and Chitty et al. The LA, Z scores, and the corresponding percentiles for an LHR of 1.0 as well as for a 60% o/e LHR (which corresponds to an LHR of 1.0 at 19 weeks) were constructed between 16-32 weeks ( Table 3 ).



TABLE 1

Percentile distributions of right lung area (mm 2 ) from 16-32 weeks’ gestational age





























































































































































































































































































































































Percentiles
Wk 0.01 0.015 0.027 0.05 1 2 2.5 5 10 15 20 25 30 35 40 45 50
16 23 23 29 34 67 77 79 91 104 112 119 125 130 134 139 143 148
17 36 37 43 50 90 101 103 118 133 143 151 158 164 170 175 180 186
18 54 55 62 70 116 130 132 150 167 179 188 196 204 210 217 222 229
19 76 77 85 94 147 163 165 185 205 218 229 239 247 254 262 268 275
20 101 102 111 122 181 199 202 224 246 261 273 284 293 301 310 317 325
21 129 130 140 152 218 237 240 265 290 306 320 332 342 351 360 369 378
22 158 159 170 183 255 277 280 308 335 353 368 381 392 402 413 422 432
23 188 188 201 215 294 317 321 351 381 400 417 431 443 454 466 475 486
24 217 218 232 247 333 358 362 394 427 448 466 481 494 506 519 529 541
25 247 247 262 278 371 398 402 437 472 495 514 530 545 557 571 582 595
26 274 275 290 308 407 436 441 478 515 540 561 578 594 607 621 634 647
27 299 300 317 335 441 472 477 517 556 583 605 623 640 654 670 683 697
28 321 322 340 360 472 505 510 552 594 622 646 666 683 698 715 729 744
29 339 340 359 380 499 533 540 584 629 658 683 704 723 739 756 771 787
30 353 354 374 396 521 558 564 611 658 689 716 738 757 774 793 808 825
31 361 362 383 406 538 576 583 633 682 715 743 766 787 804 824 840 858
32 362 364 385 410 548 589 596 648 700 735 763 788 810 828 848 866 885

Quintero. The QLI. Am J Obstet Gynecol 2011.

Derived from data of Peralta et al.


TABLE 2

Percentile distributions of lung-to-head circumference ratio from 16-32 weeks’ gestational age














































































































































































































































































































































































GA/Z score Percentiles
0.01 0.015 0.027 0.05 1 2 2.5 5 10 15 20 25 30 35 40 45 50
−3.62 −3.61 −3.46 −3.29 −2.33 −2.05 −2 −1.64 −1.28 −1.04 −0.84 −0.67 −0.52 −0.39 −0.25 −0.13
16 0.40 0.40 0.44 0.48 0.70 0.77 0.78 0.86 0.95 1.00 1.05 1.09 1.12 1.16 1.19 1.22 1.25
17 0.50 0.50 0.54 0.58 0.82 0.89 0.91 1.00 1.09 1.15 1.20 1.24 1.28 1.31 1.35 1.38 1.41
18 0.59 0.59 0.63 0.68 0.94 1.01 1.03 1.12 1.22 1.28 1.34 1.38 1.42 1.46 1.50 1.53 1.56
19 0.67 0.68 0.72 0.77 1.04 1.12 1.14 1.24 1.35 1.41 1.47 1.52 1.56 1.60 1.64 1.68 1.71
20 0.75 0.75 0.80 0.85 1.15 1.23 1.25 1.36 1.47 1.54 1.60 1.65 1.70 1.74 1.78 1.82 1.86
21 0.82 0.83 0.87 0.93 1.24 1.33 1.35 1.46 1.58 1.66 1.72 1.78 1.82 1.87 1.91 1.95 1.99
22 0.89 0.89 0.94 1.00 1.33 1.42 1.44 1.56 1.69 1.77 1.83 1.89 1.94 1.99 2.04 2.08 2.12
23 0.95 0.95 1.00 1.06 1.41 1.51 1.53 1.66 1.78 1.87 1.94 2.00 2.06 2.10 2.15 2.20 2.24
24 1.00 1.00 1.06 1.12 1.48 1.59 1.61 1.74 1.88 1.97 2.04 2.11 2.16 2.21 2.27 2.31 2.36
25 1.04 1.05 1.11 1.17 1.55 1.66 1.68 1.82 1.96 2.06 2.14 2.21 2.26 2.32 2.37 2.42 2.47
26 1.08 1.09 1.15 1.22 1.61 1.73 1.75 1.90 2.05 2.14 2.23 2.30 2.36 2.41 2.47 2.52 2.57
27 1.11 1.12 1.18 1.25 1.67 1.79 1.81 1.96 2.12 2.22 2.31 2.38 2.45 2.50 2.56 2.61 2.67
28 1.14 1.14 1.21 1.29 1.72 1.84 1.86 2.02 2.19 2.29 2.38 2.46 2.53 2.58 2.65 2.70 2.76
29 1.16 1.16 1.23 1.31 1.76 1.89 1.91 2.08 2.25 2.36 2.45 2.53 2.60 2.66 2.73 2.78 2.84
30 1.17 1.17 1.25 1.33 1.79 1.93 1.95 2.13 2.30 2.42 2.51 2.59 2.67 2.73 2.80 2.86 2.92
31 1.17 1.18 1.25 1.34 1.82 1.96 1.99 2.17 2.35 2.47 2.57 2.65 2.73 2.79 2.86 2.92 2.99
32 1.17 1.18 1.26 1.34 1.84 1.99 2.01 2.20 2.39 2.51 2.62 2.70 2.78 2.85 2.92 2.98 3.05

GA , gestational age.

Quintero. The QLI. Am J Obstet Gynecol 2011.

Derived from data of Peralta et al and Chitty et al.


TABLE 3

Lung area, percent lung area, and percentiles associated with LHR of 1.0 and fixed observed/expected LHR of 60%






































































































































































Gestational age, wk LA (mm 2 ) corresponding to LHR of 1 ( Table 1 ) Percent LA associated with LHR of 1 Percentile of LA associated with LHR of 1 ( Table 2 ) 60% Of expected LHR LA (mm 2 ) corresponding to 60% expected LHR Percent LA associated with 60% expected LHR Percentile of LA associated with 60% expected LHR
16 125 85% 14.68 0.75 74.28 50.33 1.66
17 138 74% 5.26 0.85 93.39 50.28 1.25
18 151 66% 1.81 0.94 116.89 51.13 1.02
19 164 60% 0.65 1.03 144.24 52.37 0.84
20 177 54% 0.25 1.11 174.85 53.74 0.75
21 189 50% 0.106 1.20 208.08 55.10 0.68
22 201 47% 0.05 1.27 243.28 56.37 0.64
23 213 44% 0.027 1.35 279.74 57.54 0.62
24 224 41% 0.0153 1.42 316.77 58.57 0.60
25 235 40% 0.007 1.48 353.65 59.46 0.60
26 246 38% 0.0068 1.54 389.64 60.21 0.62
27 256 37% 0.0052 1.60 424.00 60.82 0.66
28 266 36% 0.0043 1.65 455.99 61.29 0.68
29 276 35% 0.0038 1.70 484.85 61.61 0.73
30 285 34% 0.0036 1.75 509.83 61.78 0.78
31 293 34% 0.0036 1.79 530.17 61.79 0.84
32 301 34% 0.0038 1.83 545.10 61.62 0.94

LA , lung area; LHR, lung-to-head circumference ratio.

Quintero. The QLI. Am J Obstet Gynecol 2011.


The mathematical relationship between LA and HC was then analyzed. A new parameter of right lung growth was derived from this analysis.


No human subjects were involved in this research, therefore 45 CFR 46 (institutional review board approval) does not apply.




Results


Association of LHR and o/e LHR with gestational age


Table 1 displays the percentiles of right LA by gestational age (16-32 weeks). Table 2 shows the LHR percentile distributions for the gestational ages from 16-32 weeks. For each fixed percentile, there is a clear increasing trend in LHR values with advancing gestational age. Table 3 demonstrates that a fixed LHR of 1.0 ranges from the 14th percentile at 16 weeks to the .006th percentile at 26 weeks. Similarly, a fixed o/e LHR of 60% corresponds to the 5th percentile of right LA at 16 weeks, but to < .01st percentile of right LA at 26 weeks.


Derivation of the quantitative lung index (QLI)


The mathematical reason why the LHR is not independent of gestational age is the following: the lung measurement is an area (mm 2 ), whereas the HC measurement is linear (mm). Expressed mathematically, the LA follows a quadratic formula of the form:


LA(t)=u×f(t)2+v×f(t)+w[mm2]
L A ( t ) = u × f ( t ) 2 + v × f ( t ) + w [ mm 2 ]
where u , v , and w are appropriate constants, and f(t) is a function of time.


Similarly, the HC is a linear function of the form:


HC(t)=m×g(t)+b[mm]
H C ( t ) = m × g ( t ) + b [ mm ]
where m and b are appropriate constants, and g(t ) is a function of time.


If we then assume that the f(t) is very similar to g(t) , and rename f(t) as t′ :


LA(t)=u×t′2v×t′+w[mm2]
L A ( t ) = u × t ′ 2 v × t ′ + w [ mm 2 ]

HC(t)=m×t′+b[mm]
H C ( t ) = m × t ′ + b [ mm]


Therefore, the LHR is a division of 2 polynomials of different order. If we divide a polynomial of second order by a polynomial of first order, as shown below:


LHR=u×t′2+v×t′+wm×t′+b=p×t′+q+R(t′)[mm]
L H R = u × t ′ 2 + v × t ′ + w m × t ′ + b = p × t ′ + q + R ( t ′ ) [ mm]
where R(t′) is a residual that depends on t′ .


For very large values of t′ :


limt′→∞R(t′)=0
lim ⁡ t ′ → ∞ R ( t ′ ) = 0


Therefore, LHR behaves as a polynomial of first order when t′ is large, ie:


LHR≅p×t′+q[mm]
LHR ≅ p × t ′ + q ⁢ ⁡ [ mm]
where p and q are the slope and the intercept of a linear equation. Since we know that f(t) and g(t) are monotonically growing functions, then p is a positive integer, and LHR is a monotonically growing function.


When t′ approaches 0, then:


limt′→0LHR=limt′→0u×t′2+v×t′+wm×t′+b=wb=r[mm]
lim ⁡ t ′ → 0 L H R = lim ⁡ t ′ → 0 u × t ′ 2 + v × t ′ + w m × t ′ + b = w b = r [ mm]

Only gold members can continue reading. Log In or Register to continue

Related posts:

  1. Risk factors for eclampsia: a population-based study in Washington State, 1987–2007
  2. Lymphadenectomy influences the utilization of adjuvant radiation treatment for endometrial cancer
  3. Hydroxychloroquine reduces binding of antiphospholipid antibodies to syncytiotrophoblasts and restores annexin A5 expression
  4. Placental mesenchymal dysplasia

Stay updated, free articles. Join our Telegram channel
Tags:
May 25, 2017 | Posted by in GYNECOLOGY | Comments Off on The quantitative lung index (QLI): a gestational age–independent sonographic predictor of fetal lung growth

Full access? Get Clinical Tree
Get Clinical Tree app for offline access