Systems and methods for controlling breathing rate

[0001] This application claims priority under 35 U.S.C. § 119(e) from provisional U.S. patent application No. 60/837,205 filed Aug. 11, 2006 the contents of which are incorporated herein by reference.

[0002] 1. Field of the Invention

[0003] The present invention relates generally to systems and methods for controlling the breathing rate of a patient. More specifically, embodiments of the present invention are directed to controlling the application of air pressure (e.g., in a bi-level positive air pressure system) to assist a patient with reducing his/her breathing rate.

[0004] 2. Description of the Related Art

[0005] Hypertension is a significant problem in nearly all western cultures and is an underlying cause for stroke and heart attack. Termed the “Silent Killer,” hypertension affects approximately 1 in 4 Americans, and occurs with even higher prevalence in some European communities. Hypertension is also gaining recognition as a co-morbid factor in obstructed sleep apnea (OSA) patient populations, with recent studies indicating that as many as 80% of patients seeking treatment for OSA may unknowingly suffer from this disease.

[0006] Slower breathing has been shown to effectively reduce hypertension.

[0007] Indeed, a focus of techniques such as Yoga and Qigong is to slow a person's breathing to effectively relieve stress and lower blood pressure. In recent years, the physiological mechanisms underlying this response have been shown to involve changes in the autonomic nervous system through the following chain of events:

• • [0008] The process of slow breathing requires the autonomic nervous system to make more frequent adjustments in blood pressure.
  • [0009] In response to these frequent changes, the autonomic nervous system becomes more sensitive to BP changes (via the baroreflex).
  • [0010] The baroreflex becomes amplified.
  • [0011] In order to better regulate BP, a readjustment of central neural control occurs by increasing vagal activity and inhibiting sympathetic activity, both of which act to reduce BP.

[0012] Several studies support the idea of using slower breathing to reduce blood pressure in hypertensive individuals. Currently, one device on the market (RESPeRATE by Intercure™ (Fort Lee, N.J.)) is FDA approved for the treatment of hypertension. The RESPeRATE device is designed to act as a breathing “coach” by using audiovisual cues to guide the user to a slower breath rate.

[0013] The operation of this device is described, at least in part, in U.S. Pat. No. 5,800,337 to Gavish. Gavish discloses a system for modifying naturally occurring biorhythmic activity by changing at least one “non-frequency” characteristic of input to the user in response to changes in the biorhythmic activity of the user. A focus of Gavish is to effect absolute and relative durations of portions or segments of an overall biorhythmic pattern, such as breathing, independently of changes in the overall frequency of the user's biorhythm being monitored (again, e.g., breathing). The input to the user that is disclosed by Gavish includes audio, visual or tactile cues.

[0014] While still other biorhythmic feedback methodologies are known, there is nevertheless a need for different approaches to effect the breathing rate or frequency of a patient to induce, among other things, lower blood pressure, or simply induce improved periods of relaxation.

[0015] Described herein is a methodology, generally referred to as the paced breathing (PB) method, for helping patients breathe at a therapeutic rate through the use of positive pressure cues to, among other things, reduce their blood pressure.

[0016] An aspect of the invention is to employ positive air pressure (PAP) cues from a suitable device such as a Bi-Level PAP (BiPAP) device available from Respironics (Murrysville, Pa.) as a more natural proprioceptive signal to coach patients to breathe at a slow, therapeutically significant breath rate (e.g., <10 breaths per minute (BPM)). The methodology disclosed herein provides a drug-free way for some patients to reduce their blood pressure, among other physiological benefits.

[0017] In one possible implementation, a method is provided that monitors the respiration of a patient or user to determine the user's breathing or respiration frequency.

[0018] Respiration can be considered to comprise an inspiration portion or time and an expiration portion or time. During the inspiration portion of the user's respiration a HI pressure of the Bi-level PAP is supplied, and during the expiration portion, the LO pressure of the Bi-level PAP is supplied. The user's breathing frequency is then compared to a target breathing frequency. When the user's breathing frequency is greater than the target frequency, the time over which the HI pressure of the Bi-level PAP is supplied is increased, in a predetermined manner, and the time over which the LO pressure of the Bi-level PAP is supplied is adjusted. The LO pressure time may be adjusted in accordance with a fixed ratio between the HI and LO pressure times, or may be adjusted to coincide with the user's actual expiration time. Those skilled in the art will appreciate that aside from positive air pressure, one or more of the pressure cues may be positive, negative, or atmospheric.

[0019] The methodology described herein may also maintain a counter or timer to track the amount of therapy time the user has had over a predetermined period of time, e.g., over a given week. While not critical, this may indeed be an important feature since there is a reported threshold of 180 minutes of slow breathing across an 8 week period for some breathing therapies to be effective. Therapy data can be stored and reported using, e.g., well-known SmartCard technology.

[0020] Another possible implementation might include the use of a blood pressure cuff (sphygmomanometer) in conjunction with a Bi-PAP device to track and report blood pressures using the same reporting technologies found in conventional xPAP machines.

Additional Effects

[0021] It is possible that the methodology and device for inducing slow breathing described herein may have effects outside of hypertension. Similar baroreflex sensitivity changes have been observed with changes in pain, and thus the device may have hypoalgesic effects. With the associated autonomic influences, this therapy may eventually be shown to benefit additional disorders including:

• • [0022] Migraine
  • [0023] Anxiety (and Stress)
  • [0024] Restless Legs Syndrome
  • [0025] Obesity
  • [0026] Chronic Fatigue/Fibromyalgia
  • [0027] Mood Disorders
  • [0028] Slow Patterned Breathing During Birth Labor

[0029] These and other objects, features, and characteristics of the present invention, as well as the methods of operation and functions of the related elements of structure and the combination of parts and economies of manufacture, will become more apparent upon consideration of the following description and the appended claims with reference to the accompanying drawings, all of which form a part of this specification, wherein like reference numerals designate corresponding parts in the various figures. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the invention. As used in the specification and in the claims, the singular form of “a”, “an”, and “the” include plural referents unless the context clearly dictates otherwise.

[0034] Embodiments of the present invention have the capacity to reduce the breath rate of a patient using a positive pressure device. A positive pressure device is well-known and is disclosed, for example, in U.S. Pat. No. 6,105,575, hereby incorporated by reference in its entirety. In this same field, continuous positive airway pressure (CPAP) has been used for many years and has proven to be helpful in promoting regular breathing. Inspiratory positive air pressure (IPAP), of which bi-level positive air pressure (BiPAP) is a category, has also been known. In BiPAP, two levels of positive air pressure (HI and LO) are supplied to a patient. Properly controlling the timing of the HI and LO segments of BiPAP in accordance with the principles of the present invention can more effectively assist a patient in reducing his/her breathing rate.

[0035] In this regard, and in accordance with an embodiment of the present invention, the paced breathing (PB) method and related system (collectively referred to herein as “the PB method”) first measures the breath rate of the patient and then determines an optimal “machine inspiration time” (T_insp^M) and “machine expiration time” (T_exp^M) to act as a breathing “coach.” The T_insp^M and T_exp^M correspond to the application of HI or LO pressure to the patient in the context of BiPAP therapy. By setting T_insp^M and T_exp^M, the PB method may help to control or effect the breathing frequency of the patient. In control terms, the patient's breath rate is the measured variable (BR_p) and the periods of inspiration and expiration (T_insp^M and T_exp^M) are the control variables used for feedback control of breath frequency.

[0036] The PB method is preferably comprised of two levels of control, but those skilled in the art will appreciate that the methodology described herein may be implemented as a single programming module, or broken up into still greater numbers of modules or levels of control. As shown in FIG. 1, there is provided

• • [0037] 1. a Top Level Driving System (TLDS) 102 in which the target breath rate is set, and
  • [0038] 2. a Low Level Control System (LLCS) 104 that uses the target breath rate (BR_target) as input and sets a machine inspiration and expiration time (T_insp^M and T_exp^M) according to predetermined control logic. Feedback is provided to TLDS 102 through the variable BR_target Update (discussed below).

[0039] In accordance with one embodiment, and as explained below, the periods of inspiration and expiration (T_insp^M and T_exp^M) are preferably set at a fixed ratio, regardless of breath rate, such that the variable for feedback control of the patient's breathing is breath frequency. During instances when the patient may be slowing his/her breath rate more quickly than the machine's targeted breath rate, TLDS 102 updates its target breath rate via a command (i.e., BR_target Update) sent by LLCS 104.

Top Level Driving System

[0040] An overall objective of TLDS 102 is to drive the patient's breathing (BR_p) to the lowest reasonable frequency for, e.g., at least 10 minutes. TLDS 102 operates to reduce the patient's breath rate by stepping down the target breath rate (BR_target) and sending this value to the low level control system (LLCS) 104. TLDS 102 determines the rate at which the patient should be breathing and LLCS 104 responds to BR_target by setting control variables (T_insp^M and T_exp^M). FIG. 2 depicts a flow diagram showing the sequence of instructions to step down the target breath rate to assist in coaching a patient to breath more slowly.

[0041] Variables depicted in FIG. 2 are defined as follows:

• • [0042] t_s—time spent in Synchronous (S) Mode
  • [0043] BR_target(k)—target breath rate
  • [0044] BR_p(k)—average patient breath rate
  • [0045] BR(k)—instantaneous breath rate
  • [0046] t—time spent at current breath rate
  • [0047] t_h—total therapy time
  • [0048] X—sampling period in S-mode
  • [0049] y—max time allowed at current breath rate
  • [0050] T_insp—patient's time of inspiration
  • [0051] T_e—patient's time of expiration

[0052] To step down the patient's breath rate, a sequence of steps 200 of TLDS 102 operates LLCS 104 in a synchronous mode (S-Mode) for X amount of time (i.e., T_insp^M=T_insp and T_e^M=T_e) and the patient's average breath rate BR_p(k) is monitored, as shown at steps 202 and 204. That is, the machine is set to follow precisely the patient's breathing pattern. At step 206, once the S-mode operation has been completed, TLDS 102 reduces the target breath rate by a breath rate unit (BRU) below the current average breath rate measured during S-mode and resets a therapy timer t_h to determine a next target breath rate update and the conclusion of therapy time. At step 208 a breath count k is incremented at the end of expiration. Thereafter, at step 210, TLDS 102 resets a timer t, which represents the time spent at the current breath.

[0053] At the end of each breath, TLDS 102 determines, at step 212, whether or not the last breath (BR(k)) was at or below (the exemplary) 10 BPM (breaths per minute). If so, then at step 214 the breath period is added to the therapy timer, t_h. If the therapy timer has not reached a total time of 10 minutes or more as shown at step 216, then, at step 218, TLDS 102 polls the LLCS to see if there is an updated target breath rate. If so, then BR_target is updated at step 220 and a new cycle is started. Otherwise TLDS 102 determines, at step 222 if it is time to update the target breath rate by polling timer t (which indicates the time spent at the current breath rate). More specifically, if the time to update target breath has been reached (i.e., y amount of time has elapsed since the last time target breath rate was updated, or t≧y time), then the target breath rate is again updated according to the following rules implemented by step 224:

• • [0054] 1. If the patient is breathing more slowly than the machine (i.e., the current target breath rate is greater than or equal to the patient's average breath rate), then the new target breath rate is updated to the average breath rate minus one BRU (step 226).
  • [0055] 2. If the patient is breathing faster than the machine, then the target breath rate is set to the patient's breath rate (step 228). If the values are equal, step 226 is performed to reduce the user's breath rate. The target is set to the patient's rate preferably only when the patient rate is faster than the machine.

[0056] More simply put, if the patient is breathing more slowly than the machine, the machine is adjusted to push the patient's breathing to a slower pace. Otherwise, if the patient is breathing faster than the machine, the machine “backs up” and adjusts itself to the breathing rate of the patient.

[0057] Finally, if at the end of the breath no action is required (i.e., none of the cases above were executed), then the timer t is reset, and the TLDS waits for the end of another breath. If at step 212 the patient's instantaneous breath rate was not less than 10 BPM, then the breath period is not added to the total therapy time t_h and the process continues with step 216. If the total therapy time t_h is greater than, e.g., 10 minutes as depicted at step 216, then the patient is deemed to have had a breath rate at the desired frequency for at least 10 minutes and the process returns to S-mode, as illustrated by step 230.

Low Level Control System

[0058] The target breath rate generated by TLDS 102 is sent to LLCS 104 where it is used to determine the T_insp^M and T_e^M to be used on the current breath cycle. To determine the value of T_insp^M and T_e^M, LLCS 104 uses standard, well-known, integral control equations with anti-windup mechanisms in addition to a state machine mode that chooses adequate control values to provide the maximum comfort to the patient.

[0059] FIG. 3 illustrates the components governing the control mechanism in block diagram form. Specifically, depicted are:

• • [0060] 1. Breath rate (BR) low pass filter 302,
  • [0061] 2. Controller 304,
  • [0062] 3. State Machine Mechanism 306, and
  • [0063] 4. Lung Dynamics 308.

[0064] Breath rate (BR) low pass filter 302 determines an average breath rate for the last n breaths (by monitoring lung dynamics 308), and provides controller 304 with a measure of the respiratory state of the patient. Controller 304 estimates {circumflex over (T)}_insp^M that seems adequate for the current breath using the reading of the filter breath rate and the target breath rate. Finally, state machine 306 takes the output of controller 304 as a reference and uses it to compute the actual machine periods of inspiration and expiration states, T_insp^M and T_e^M, respectively. The detailed equations that express the behavior of these blocks are described below.

[0065] To determine the value of {circumflex over (T)}_insp^M, the PB method uses a breath rate controller whose equation resembles an integral controller (see eq. (1) and eq. (2).) An objective of controller 304 is to increase the machine estimated inspiration time {circumflex over (T)}_insp^M as the error (e(k)) between the current target breath rate and the current low pass filtered patient's breath rate increases as follows:
e(k)=BR_target(k)−BR_p(k),   (1)
{circumflex over (T)}_insp^M(k)=K_ie(k)+T_insp^M(k−1),   (2)

[0066] In these equations, k is the current breath number, e(k) is the current error signal between current target breath rate BR_target(k) and the patient's current breath rate after filtering BR_p(k), and {circumflex over (T)}_insp^M(k) is the estimated machine time of inspiration to be used by state machine 306.

The BR Low Pass Filter

[0067] In one possible embodiment, breath rate low pass filter 302 uses a single FIR digital filter to determine an average breath rate. An exemplary equation governing the behavior of this block is:

[0068] $\begin{array}{cc}{\mathrm{BR}}_p\left(k\right)=\sum _{m=0}^N{b}_m\left(\mathrm{BR}\left(k-m\right)\right).& \left(3\right)\end{array}$

[0069] Where BR_p(k) is the low pass filter signal and BR(k)s are the instantaneous signal samples at breath k. And, b_m is a weighting factor. For instance, b_m may be 1/N for a simple moving average. Alternatively, b_m may be a weighted average.

[0070] The State Machine

[0071] State Machine 306 determines the actual values of machine inspiration and expiration time to be provided to the patient. These values are determined as follows:
T_insp^M(k)=max({circumflex over (T)}_insp^M(k),T_insp),   (4)
{circumflex over (T)}_e^M(k)=(R)T_insp^M(k), and   (5)
T_e^M(k)=min({circumflex over (T)}_e^M(k),T(k)−T_insp^M(k)).   (6)

[0072] In equations (4), (5) and (6), k represents the current breath, {circumflex over (T)}_insp^M(k) and {circumflex over (T)}_e^M(k) are the estimated machine inspiration and expiration time, T_insp(k) is the patient's inspiration time, T(k) is the patient's periodic time (i.e., period of given entire breath) and T_insp^M(k) and T_e^M(k) represent the actual machine inspiration and expiration time, respectively.

[0073] From equation (4) it is noted that the machine actual inspiration time is determined by the longest time between the controller estimated time and the patient's inspiration time. This mechanism provides a way to encourage patients to elongate their inspiration time. As for the patient's expiration time in equation (6), it is noted that state machine 306 will end expiration based on the shortest time of estimated machine time of expiration ({circumflex over (T)}_e^M(k)) or patient's time to expiration (T(k)−T_insp^M(k)). The machine estimated expiration time is, as mentioned before, calculated based on a fixed ratio (R) of the machine time of inspiration T_insp^M(k) (see eq. (5).)

[0074] To better understand the dynamics of equations (4)-(6), several scenarios are illustrated in FIGS. 4A-C. In these Figures, the graphs represent a patient breathing at 10 BPM, with T_insp(k)=2 secs and T_e(k)=4 secs.

[0075] Case A in FIG. 4A, illustrates when {circumflex over (T)}_insp^M(k)<T_insp(k), making T_insp^M(k) equal to T_insp(k)=2 secs. This case also illustrates when {circumflex over (T)}_e^M(k)>T_e(k) and therefore T_e^M(k) equals T_e(k)=4 secs. In case A, the machine is substantially perfectly aligned with the patient breathing.

[0076] Case B of FIG. 4B illustrates when {circumflex over (T)}_insp^M(k)>T_insp(k). In this case, T_insp^M(k)={circumflex over (T)}_insp^M(k)=2.5 secs, and T_insp(k)=2.0 secs. Note that the effect is that the patient will be provided IPAP HI pressure for a longer period of time as illustrated in the graph. Finally, case C of FIG. 4C illustrates when {circumflex over (T)}_e^M(k)<T_e(k). In this case, T_e^M(k) equals {circumflex over (T)}_e^M(k)=3.5 secs and the machine will initiate another inspiration period before the patient will actually initiate inspiration.

[0077] Note that the essence of the LLCS is to choose between two possible frequencies f₁ or f₂ on a breath-by-breath basis. The variable f₁ is defined by the patient's breath rate (see cases A and B of FIGS. 4A and 4B,) and f₂ is defined as a fixed lower frequency determined by fixed ratio (R) (as shown in case C of FIG. 4C.)

[0078] When the preset frequency f₂ is chosen, the patient's breath rate f₁ is slower than the device's current target frequency. In this case, it is desirable that the machine reduce its coaching frequency f₁ via the TLDS. In this scenario, state machine 306 sends an update to the target breath rate below the current patient's breath rate (BR_p(k)) as shown in FIG. 1 and FIG. 2.

[0079] Based on the foregoing, those skilled in the art will appreciate the novel aspects of the embodiments of the PB method. Still other benefits of the PB method are described below.

[0080] As explained in detail above, the PB method uses pressure cues to coach the user's breathing pattern to a targeted frequency. As such, devices intended to deliver gases other than room air can also be configured to operate in a similar manner. Examples of such devices include oxygen delivery devices, anesthesia devices, fresh air respirators, and respiratory (or other) drug delivery devices. It should also be appreciated that the cues may be positive pressure, negative pressure, or atmospheric. In addition, the inventors also contemplate that the unique features of the present invention may be incorporated into a stand-alone device which utilizes any humanly perceptible stimulus to cue the user. In addition to pressure based cues as described above, the cues may also be tactile, visual, or sonic.

[0081] Also, because the PB method involves tracking patient synchronization with the device delivering pressure cues, it is possible to employ the method to track sleep onset. If, for example, after a period of successfully lowering breath rate using pressure cues from the device, a patient is no longer able to synchronize with the PB method (despite several iterations attempting to do so), the device can assume the user is asleep and no longer cognitively able to interact with the breathing coach. In such cases, the machine can provide an accurate assessment of sleep onset time. This feature may represent a valuable commodity to the health care provider since it can provide a more accurate assessment of true sleep time with the devices versus merely “machine run time.”

[0082] As noted previously, the PB method incorporates a breathing coach that actively works to reduce the user's breath rate. There is both scientific and anecdotal knowledge supporting the notion that such measures can reduce stress and induce a more relaxed state. Qigong, yoga, and other deep-breathing exercises are used by millions for stress relief and many other purposes. The scientific merit of such behavior modification has become well established in recent years.

[0083] For similar reasons, the PB method can be used in the treatment of chronic pain syndrome. The mechanisms through which the device works for indication are not fully understood, but may be related to the changes in baroreflex sensitivity that mediate the noted effects on hypertension.

[0084] Also for similar reasons, the PB method may be use in lieu of or in addition to anxiolytics to combat anxiety.

[0085] Previous evidence has suggested that deep breathing exercises can produce effective reductions in cholesterol associated with hypertension and stress. It is thought that the mechanisms underlying these effects may be related to central changes associated with baroreflex sensitivity following several weeks of deep breathing intervention. Whether the changes in cholesterol levels are the result of reduced hypertension and stress levels or some parallel phenomena are unknown.

[0086] The PB method may also be effective in weight loss programs.

[0087] Likewise, the PB method may be helpful in the detection and treatment of Attention Deficit Hyperactivity Disorder (ADHD).

[0088] Also, the PB method may be helpful in the detection and treatment of Panic Disorder.

[0089] In addition, because of the propensity of the techniques of the PB method to induce relaxation and reduce stress, it may also be desirable to monitor the level of “relaxation” felt by the user. Such measurements might include:

• • [0090] Sympathetic tone (tonometry)
  • [0091] Non-invasive Blood Pressure (via cuff)
  • [0092] Plethysmography
  • [0093] Pulse Transit Time
  • [0094] Galvanic Skin Response
  • [0095] Pupillary diameter
  • [0096] Muscle Sympathetic Nerve Activity (MSNA)

[0097] Although the invention has been described in detail for the purpose of illustration based on what is currently considered to be the most practical and preferred embodiments, it is to be understood that such detail is solely for that purpose and that the invention is not limited to the disclosed embodiments, but, on the contrary, is intended to cover modifications and equivalent arrangements that are within the spirit and scope of the appended claims.